Next Page: 10000

          Distilled News      Cache   Translate Page      
Toward an AI Physicist for Unsupervised Learning We investigate opportunities and challenges for improving unsupervised machine learning using four common …

Continue reading


          New Quantum Machine Learning Algorithm Could Crush Big Data      Cache   Translate Page      
Read Zayan Guedim explain how a new hybrid system combines classical machine learning techniques with quantum computation to optimize the management of the power grid on Edgylabs blog : Strategically deployed across the grid, Phasor Measurement Units (PMUs) are sensor devices that measure currents and voltages at a particular time. PMUs generate huge amounts of data that help […]
          SR. ANALYST - TransUnion - Chicago, IL      Cache   Translate Page      
Segmentation, regression, clustering, survival analysis, and machine learning). Our culture encourages our people to hone current skills and build new...
From TransUnion - Fri, 14 Sep 2018 16:14:34 GMT - View all Chicago, IL jobs
          Algorithms define our lives      Cache   Translate Page      
also btw... Revised and Extended Remarks at 'The Rise of Intelligent Economies and the Work of the IMF' - "The dirty secret of the field, and of the current hype, is that 90% of machine learning is a rebranding of nonparametric regression."
          T-SQL Tuesday – Non-SQL Server Technologies      Cache   Translate Page      

MJTuesday

So, this month’s T-SQL Tuesday topic is to think about a non-SQL Server technology that we want to learn.

For me, I’m going to pick machine learning.

As a DBA, I’ve always looked at machine learning as a thing for the BI guys.  I’m a DBA after all why do I care about that?

Well, my attitude has changed somewhat recently.  This little change all started when I listened to Alex Whittles’ keynote talk at Data Relay.  He presented a demo where a computer program used Python (I’m already a huge fan of Python in SQL Server as you may know) and SciPy (a machine learning, data sciencey type module) to play and learn a game.  Alex demonstrated how, over time his robot was able to increase it’s score through machine learning algorithms.

WOW, myself and Adrian looked at each other as a little light bulb come on over our heads.  For the rest of the conference I attended a number of sessions that I wouldn’t normally attend, stuff for the BI guys.  A great session from Terry McCann and an interesting one from Simon Whiteley really got the creative juices flowing.  Could the DBA use this technology to model things like performance trends, predict capacity and answer that question that we’re always asked, “have we got room on the SQL Server for just one more DB?”.

So where do I go from here?  My first port of call is going to get my head around Python, I’ve got a background in C programming to that shouldn’t be too difficult.  Once I’m happy with that, it’ll be a case of hitting the blogs, courses, books and anything else that I can get my hands on to help understand the strange mysteries that are Machine Learning.

Where can I go with this?  As DBAs, we’ve got a ton of data available to us in DBVs, Query Store, etc.  Wouldn’t it be great if we could hook a little robot into all that and start building up models of how our servers behave.  Keep an eye out for the inevitable blog post that are going to come out of it.

 


          Machine Learning Algorithm Developer - ROSS Recruitment - North York, ON      Cache   Translate Page      
MS or PHD in Computer Science. Our client is a well-established leader in online sports gaming with a Technical Centre of Excellence at Yonge and Sheppard....
From ROSS Recruitment - Sat, 29 Sep 2018 03:52:35 GMT - View all North York, ON jobs
          Machine Learning Algorithm Dev - $1000 ref fee - ROSS Recruitment - North York, ON      Cache   Translate Page      
MS or PHD in Computer Science, specializing in Machine Learning is desirable. Our client is a well-established leader in online sports gaming with a Technical...
From ROSS Recruitment - Sat, 29 Sep 2018 04:00:33 GMT - View all North York, ON jobs
          Remote Data Scientist      Cache   Translate Page      
A staffing firm needs applicants for an opening for a Remote Data Scientist. Core Responsibilities of this position include: Building state of the art, scalable, and self-learning systems Training and tuning a variety of machine learning models Performing data and error analysis Must meet the following requirements for consideration: 5 years of experience with complex, self-directed data analysis 1 year of practical work experience using cloud services such as AWS Experience in at least one Statistical Modeling, Machine Learning, Predictive Analytics, Data Visualizations Hands-on with experiments and execution Experience executing document clustering Experience in analyzing English language text-based data sets
          Ted Cruz is still using a blacklisted Cambridge Analytica app developer      Cache   Translate Page      

In his re-election campaign’s final hours, Senator Ted Cruz (R-TX) is still deploying a smartphone app created by a software team at the heart of the Cambridge Analytica controversy.

The app, Cruz Crew, was developed by AggregateIQ, a small Canadian data firm that was for years the lead developer used by the infamous data analytics consultancy that made headlines last spring for harvesting user data on millions of unsuspecting Facebook users while working for the Trump campaign. Since that firm’s demise, AggregateIQ has become one focus of an international investigation into alleged data misdeeds during the 2016 Brexit campaign, and is the first company to be targeted by regulators under Europe’s new data privacy law.

The Cruz Crew app’s login screen. The app’s Facebook login was finally removed in June. [Image: Google Play]
Both Cruz Crew as well as an app for Cruz’s presidential campaign in 2016 share an interconnected history of developers and clients linked to Cambridge Analytica, its British affiliate SCL Elections, and architects of the Republican Party’s recent digital efforts. Part of a group of apps presented as walled-garden social networks for political supporters, the software helps campaigns collect voter data and microtarget messages.

In April, Facebook announced it had suspended AggregateIQ over its possible improper access to the data of millions of Facebook users. But over a dozen apps made by AggregateIQ remained connected to Facebook’s platform until May and June, when Facebook belatedly took action against them.

A Facebook spokesperson told Fast Company that it was still investigating AIQ’s possible misuse of data, amid an ongoing investigation by Canadian prosecutors. The Cruz campaign did not respond to requests for comment.

David Carroll, a professor at Parsons School of Design at the New School in New York, who has brought a legal challenge against SCL and Cambridge Analytica for release of his voter data profile, said Cruz’s continued relationships with AggregateIQ highlighted problems with the use of data by a growing ecosystem of partisan election apps and databases. The risks are particularly high, he said, when the vendors are combining data from multiple sources and processing Americans’ data overseas.

“Despite the Cambridge Analytica fiasco, it seems that the Republican data machine is still a shadowy network that includes international operators, tangled up with vendors under intense scrutiny for unlawful conduct in multiple jurisdictions,” he said. “I don’t understand why Republicans don’t insist on working with domestic tech vendors and technologists who are U.S. citizens.”

The Cruz-Cambridge Analytica connection

During the 2016 race, a U.S.-based software firm named Political Social Media, but better known as uCampaign, was credited as developer and publisher for the official “Ted Cruz 2016” presidential primary app. At the time, the app achieved modest notoriety as a somewhat novel data collection tool– appearing alongside Cambridge Analytica under headlines like, “Cruz App Data Collection Helps Campaign Read Minds of Voters”–with the app colloquially referred to in the press as “Cruz Crew.”

As in 2016, the 2018 Cruz re-election campaign relies on constant polling and voter modeling to understand and target mainstream conservatives in Texas. Cruz and his Democratic challenger Beto O’Rourke, who has repeatedly brought up Cambridge Analytica during the campaign and has refused to use big data analytics, have both heavily invested in social media. The media blitz hasn’t been cheap: According to data from the Center for Responsive Politics, the candidates in the 2018 Texas Senate race have set the all-time record for most money spent in any U.S. Senate election.

As part of its digital push, the Cruz campaign rolled out a new app, officially named “Cruz Crew,” which awards points to users for tweeting pro-Cruz messages, volunteering, and taking part in other campaign activities. On the app’s pages in the Google and Apple stores, AggregateIQ is not mentioned, but its name is visible as the developer in the app URL and in internal code. The app’s publisher is listed as the political marketing agency WPA Intelligence, or WPAi.

Chris Wilson, WPAi’s founder and chief executive, is a veteran GOP pollster who previously worked for George W. Bush and Karl Rove. WPAi’s past campaign successes include a trio of high profile Tea Party-cum-Freedom Caucus sympathizer senators: Cruz, Mike Lee (R-UT), and Ron Johnson (R-WI). By far, however, Cruz has been WPA’s biggest political client in the U.S. Between his bids for senator and president, Cruz campaign committees have paid out over $4.3 million to Chris Wilson’s firm since 2011.

As the director of research, analytics and digital strategy for Cruz’s 2016 presidential campaign, Wilson oversaw a large data team that included Rebekah Mercer and Steve Bannon’s Cambridge Analytica. Rebekah’s father, Robert Mercer, footed the $5.8 million bill for Cambridge Analytica by doubling that amount in donations.

Wilson and the Cruz team have repeatedly said that Cambridge Analytica represented to the campaign that all of the data it had was legally obtained. They also claimed that Cambridge did not deliver the results expected of them, neither through their much-discussed psychographics work nor through an important piece of software called Ripon.

In schematics, Ripon was drawn up as an all-in-one campaign solution to manage voter data collection, ad targeting, and street canvassing. According to files retrieved by computer security analyst Chris Vickery, Ripon was intended to tap into something called “the Database of Truth.” Documents revealed that the Truth project “integrates, obtains, and normalizes data from disparate sources,” beginning with the Republican National Committee’s Data Trust database, combined “with state voter files, consumer data, third-party data providers, historical WPA survey, and projects and customer data.”

Despite being a deliverable promised by Cambridge Analytica, the work on Ripon was outsourced to AggregateIQ. More recently, WPAi hired the firm to develop and manage the software for Cruz Crew, along with its two other currently available apps: one for Texas Governor Greg Abbott’s re-election campaign, and one for Osnova, a Ukrainian political party dedicated to the long-shot presidential aspirations of its oligarch founder, Serhiy Taruta.

In the 2018 race, WPAi and the Cruz campaign have said Cruz’s effort isn’t using new Cambridge Analytica-style “psychographic” modeling, but it is using social media data for specific targeting, and relying on previous campaign data. “We use social data to ID voter groups in our core universes,” WPA’s Chris Wilson previously told Fast Company. “A lot of those are 2016 voters who we know are persuaded by specific messages.”

Cruz Crew and TedCruz.org currently share a privacy policy has barely changed since late 2015, when Cambridge Analytica and uCampaign were Cruz vendors. In both cases, the policy states that the campaign may “access, collect, and store personal information about other people that is available to us through your contact list,” match the info to data from other sources, and “keep track of your device’s geographic location.”


Related: How Ted Cruz plans to beat Beto O’Rourke: Play it simple


Beyond the existing campaign app, however, AIQ’s current involvement in the Cruz campaign’s data management and software development is unknown. A report by the New York Times last month found that when users shared their friends’ contact information with the Cruz app, that data was still being sent to AggregateIQ domains.

Wilson told the Times that his company, not AggregateIQ, received and controlled app users’ information. Representatives for AggregateIQ did not immediately respond to a request for comment, and WPAi did not respond to questions about the data firm.

Intelligence quotient

AIQ, founded in 2013 in Victoria, British Columbia, is currently under investigation in the U.K. and its homebase of Canada for electoral impropriety during the Brexit Leave campaign. The company’s name has come up repeatedly in parliamentary testimony for its alleged campaign finance and data protection misdeeds in connection with the parent company of Cambridge Analytica.

“Concerns have been raised about the closeness of the two organizations including suggestions that AIQ, [SCL Elections, and Cambridge Analytica] were, in effect, one and the same entity,” stated a recent report by the U.K.’s Information Commissioner’s Office.

In testimony to a U.K. parliamentary committee, former Cambridge Analytica executive Brittany Kaiser said that AggregateIQ was the exclusive digital and data engineering partner of SCL, the British parent affiliate of Analytica.

“They would build our software, such as a platform that we designed for Senator Ted Cruz’s campaign,” she said. “That was meant to collect data for canvassing individuals who would go door-to-door collecting and hygiening data of individuals in those households. We also had no internal digital capacity at the time, so we did not actually undertake any of our digital campaigns. That was done exclusively through AggregateIQ.”

AIQ founders Zack Massingham and Jeff Silvester had been brought into the fold a year prior by their friend Christopher Wylie, then an SCL employee, who blew the whistle on the firm’s practices earlier this year. According to Wylie, the founders registered their company in their hometown of Victoria as a result of an SCL contract, which subsequently led to political work in the Caribbean.

After the two firms first made contact in August 2013, while SCL was performing its first American political work in the Virginia gubernatorial race, AIQ designed solutions for deployment in campaigns under SCL’s supervision in Trinidad and Tobago. Part of the intent, according to records obtained by the Globe and Mail, was to harvest the internet histories of up to 1.3 million civilians in order to more accurately model their psychographics for message targeting.

In December 2013, an SCL employee proposed requesting the data from the country’s internet provider by posing as academic researchers, while seeking to tie internet addresses to billing addresses, without naming customers. In response, AIQ CEO Massingham replied by email that he could use every bit of data they could get. “If the billing addresses are obfuscated, we’ll have a difficult time relating things back to a real person or household,” he wrote. It remains unknown if that data was obtained.


Related: How Cambridge Analytica fueled a shady global passport bonanza


The primary work AIQ performed was to design software that could be used to motivate volunteers, canvassers, and voters. This software concept was repeated for multiple clients, including Petronas, an oil company that sought to influence voters in Malaysia.

Campaign software developed by AIQ was used by Cambridge Analytica in U.S. elections and for clients like the oil giant Petronas. [Image: SCL]

AggregateIQ’s work across the pond

During the U.K.’s Brexit campaign in 2016, Vote Leave hired AIQ to place online ads, with AIQ paying for all 1,034 Facebook ads run by the campaign. AIQ’s services were also retained to develop and administer a piece of software that Vote Leave executives, including chief technology officer and former SCL employee Thomas Borwick, later credited with a large portion of the campaign’s success.

Vote Leave campaign director Dominic Cummings wrote an extensive blog post about the project, called the Voter Intention Collection System (VICS).

“One of our central ideas was that the campaign had to do things in the field of data that have never been done before,” Cummings wrote. “This included a) integrating data from social media, online advertising, websites, apps, canvassing, direct mail, polls, online fundraising, activist feedback . . . and b) having experts in physics and machine learning do proper data science in the way only they can, i.e. far beyond the normal skills applied in political campaigns.”

As the voter-facing front end for the Leave campaign data team, uCampaign was brought in and paid by AIQ to deliver the smartphone apps that helped to gather users’ cell numbers, email addresses, phone book contacts, and Facebook IDs for integration, exactly as it had done during the previous months for the Cruz 2016 campaign. Just as in that case, the app collected voter information for use in AIQ tools.

“We could only do this properly if we had proper canvassing software,” Cummings wrote. “We built it partly in-house and partly using an external engineer who we sat in our office for months.”

AIQ’s Zach Massingham repeatedly flew to the U.K. as his company was paid hundred of thousands of pounds for its Vote Leave work in 2016 after a series of transactions between several campaigns that Canadian officials have questioned as “money laundering” and British authorities are investigating as criminal offenses. Nonetheless, after the referendum, Cummings released an open-source version of VICS code on Github for future micro-targeters to use.

In early 2018, one of Vote Leave and SCL vet Thomas Borwick’s handful of data firms, Kanto, was hired to do canvassing and social media work during the Irish abortion referendum. Anti-abortion activist groups also contracted uCampaign to build two separate apps, which alarmed campaign finance and privacy watchdogs and led to a ban on internet advertising.

As with uCampaign, which has also made apps for the likes of Donald Trump and the NRA, AIQ’s smartphone apps were designed to gather information via Facebook Login, a tool offered by Facebook to streamline user registration across the internet. Though Facebook tightened some restrictions this year as a direct response to the Analytica flare-up, Login has allowed third-party developers to gain access to a wide range of Facebook account information about registered users.

As part of its investigation into Cambridge Analytica and its affiliates, on April 7, Facebook said that it had suspended AIQ, effectively ending its ability to deploy Facebook Login. However, security researcher Chris Vickery discovered that AIQ’s access to the Facebook platform was still active as of May 17. Additionally, he found, AIQ had already collected info on nearly 800,000 Facebook account IDs in a database, with many matched to addresses and phone numbers. Facebook removed more AIQ apps two weeks later, but it was not until June 19 that the Facebook Login feature was removed from the apps for Cruz, Osnova, and Abbott.

In written testimony to Parliament, AIQ chief technology officer Jeff Silvester, who visited British prime minister Theresa May’s office with Massingham in the weeks after the Brexit vote, explained the history of the relationship between SCL and AIQ, which began in late 2013.

After building a “customer relationship management (CRM) tool” for SCL in Trinidad and Tobago, AIQ created “an entirely new CRM tool” for the 2014 U.S. midterm elections. “SCL called the tool Ripon,” Silvester wrote. AIQ was then required to transfer all software rights to SCL before working “with SCL on similar software development, online advertising, and website development” in support of Cambridge Analytica’s work for the Ted Cruz 2016 campaign.

A referral from “an acquaintance who was working with Vote Leave” led to AIQ being hired by Vote Leave in April 2016, the day before the campaign was designated as the official Leave organization.

[Photo: Stock Catalog]
This past May, after questioning the legality of AggregateIQ founder Zach Massingham’s work on British soil while developing VICS, parliamentary committee chair Damian Collins asked Silvester about AIQ’s recent work for WPAi.

Silvester explained, “They sell their software that we create for them to whomever they like, and we just simply support that work.”

In March, WPAi CEO Chris Wilson told Gizmodo that he had almost no knowledge of the controversy surrounding AIQ, despite their work for the Cruz 2016 campaign. “I would never work with a firm that I felt had done something illegal or even unethical,” he said. The firm’s work for WPA was the result of a competitive bidding process, he said, and AIQ “offered us the best capabilities for the best price.”

Leaving the nest

In February 2017, a story on the Politico Pro website announced Archie, WPA Intelligence’s new piece of software for 2018 campaigns. The software goes by a nickname used by Texas Governor Greg Abbott’s political team, referring to Archimedes, the Greek mathematician who said, “Give me a lever and I can move the world.”

A diagram describing Archimedes, WPAi’s new campaign software [Image: WPAi]
“The program allows campaigns to work across all formats and vendors to collect data in one place,” the article said, and campaign staffers “will be able to use the app to generate models, target audiences, cut lists, and produce data visualization tools to make strategic decisions.”

From that description, Archie sounded very much like AIQ’s Ripon and VICS all-in-one campaign solutions. AIQ’s smartphone app for WPAi client Greg Abbott first appeared on Google Play and Apple’s iOS Store three months later, in May 2017.

Archie’s predictive modeling of Texan voters “yielded approximately 4.5 million individual targets for turnout efforts,” according to WPAi. That helped the Abbott campaign win the 2018 Reed award for Best Use of Data Analytics/Machine Learning in Field Program. In attendance at the March ceremony were representatives from Cambridge Analytica, which was nominated for Best Use of Online Targeting for Judicial Campaign.

Three weeks after the Reed awards, Christopher Wylie’s whistleblower account in the Observer were splashed across the world’s front pages. By the following month, SCL and Analytica were claiming bankruptcy, and AIQ’s cofounders were appearing at Canadian Parliament and dealing with its suspension from Facebook as developers.

In June, a week before AIQ’s WPA apps finally removed Facebook Login, Silvester appeared before Canadian Parliament for a second time, where he was admonished by Vice Chair Nathaniel Erskine-Smith, who remarked, “Frankly, the information you have provided is inadequate.” After being threatened with a contempt charge for excusing himself from sworn testimony with a one-line doctor’s note, Massingham later spoke with the committee via audio-only link from his lawyer’s office.

In July, AggregateIQ was served with the U.K.’s first-ever enforcement notice under the EU’s new General Data Protection Regulation, known as GDPR. The U.K.’s Information Commissioner’s Office subjected AIQ to millions in fines if it did not “cease processing any personal data of U.K. or EU citizens obtained from U.K. political organizations or otherwise for the purposes of data analytics, political campaigning, or any other advertising purposes.”

After AIQ appealed the order, it was merely mandated to “erase any personal data of individuals in the U.K.,” though it was found to have “processed personal data in a way that the data subjects were not aware of, for purposes which they would not have expected, and without a lawful basis for that processing.”

As Ted Cruz wraps up his campaign, he continues to outsource part of his voter data harvesting to a foreign firm that has been blacklisted by Facebook and British and European regulators. The total data amassed through apps like Cruz Crew and projects like Ripon and Archimedes remains unknown, but they raise concerns that Cruz acknowledged when he launched his presidential campaign at Liberty University in March 2015. “Instead of a government that seizes your emails and your cell phones,” he said, “imagine a federal government that protected the privacy rights of every American.”


Jesse Witt (@witjest) is an independent researcher, writer, and filmmaker.

With additional reporting by Alex Pasternack.


          How AI, machine learning is transforming customer interactions at Schneider Electric      Cache   Translate Page      
How AI, machine learning is transforming customer interactions at Schneider ElectricHerve Coureil, chief digital officer, Schneider Electric, discusses how AI, machine learning has allowed the energy automation giant to derive the sentiment of consumers who interact with them, as well mining as much data as possible so future strategies can resonate with the broader community. This interview is part of Beyond Technology, a series powered […]
          SSIS Best Online Training (KOLKATA)      Cache   Translate Page      
SQL School is one of the best training institutes for Microsoft SQL Server Developer Training, SQL DBA Training, MSBI Training, Power BI Training, Azure Training, Data Science Training, Python Training, Hadoop Training, Tableau Training, Machine Learning Training, Oracle PL SQL Training. We have been providing Classroom Training, Live-Online Training, On Demand Video Training and Corporate trainings. All our training sessions are COMPLETELY PRACTICAL. SSIS COURSE DETAILS - FOR ONLINE TRAINING: SQL ...
          Machine Learning Algorithm Developer - ROSS Recruitment - North York, ON      Cache   Translate Page      
Our client is a well-established leader in online sports gaming with a Technical Centre of Excellence at Yonge and Sheppard. They have engaged ROSS to help...
From ROSS Recruitment - Sat, 29 Sep 2018 03:52:35 GMT - View all North York, ON jobs
          Machine Learning Algorithm Dev - $1000 ref fee - ROSS Recruitment - North York, ON      Cache   Translate Page      
Python and/or .NET (C# or VB) is a plus. Our client is a well-established leader in online sports gaming with a Technical Centre of Excellence at Yonge and...
From ROSS Recruitment - Sat, 29 Sep 2018 04:00:33 GMT - View all North York, ON jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Seattle, WA      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 07 Sep 2018 02:02:14 GMT - View all Seattle, WA jobs
          Sr. Associate, ML Pipelines for AI Consultant - KPMG - McLean, VA      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 26 Oct 2018 15:22:01 GMT - View all McLean, VA jobs
          Sr. Associate, AI in Management Analytics Consultant - KPMG - McLean, VA      Cache   Translate Page      
Ability to apply statistical, machine learnings, and artificial intelligence techniques to achieve concrete business goals and work with the business to...
From KPMG LLP - Sat, 29 Sep 2018 15:21:53 GMT - View all McLean, VA jobs
          Data Scientist - Deloitte - Springfield, VA      Cache   Translate Page      
Demonstrated knowledge of machine learning techniques and algorithms. We believe that business has the power to inspire and transform....
From Deloitte - Fri, 10 Aug 2018 06:29:44 GMT - View all Springfield, VA jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Dallas, TX      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 07 Sep 2018 02:02:14 GMT - View all Dallas, TX jobs
          Associate, Machine Learning AI Consultant, Financial Services - KPMG - Dallas, TX      Cache   Translate Page      
Broad, versatile knowledge of analytics and data science landscape, combined with strong business consulting acumen, enabling the identification, design and...
From KPMG LLP - Fri, 07 Sep 2018 02:02:14 GMT - View all Dallas, TX jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Philadelphia, PA      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 14 Sep 2018 08:38:34 GMT - View all Philadelphia, PA jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - New York, NY      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 14 Sep 2018 08:38:34 GMT - View all New York, NY jobs
          Associate, Machine Learning AI Consultant, Financial Services - KPMG - New York, NY      Cache   Translate Page      
Broad, versatile knowledge of analytics and data science landscape, combined with strong business consulting acumen, enabling the identification, design and...
From KPMG LLP - Fri, 14 Sep 2018 08:38:34 GMT - View all New York, NY jobs
          Data Scientist: Medical VoC and Text Analytics Manager - GlaxoSmithKline - Research Triangle Park, NC      Cache   Translate Page      
Strong business acumen; 2+ years of unstructured data analysis/text analytics/natural language processing and/or machine learning application for critical...
From GlaxoSmithKline - Fri, 19 Oct 2018 23:19:12 GMT - View all Research Triangle Park, NC jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Charlotte, NC      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Wed, 17 Oct 2018 08:49:16 GMT - View all Charlotte, NC jobs
          NXP Doubles Down on Machine Learning at the Edge      Cache   Translate Page      
From Arm to Google to HPE to Microsoft, many tech heavyweights have worked to streamline machine learning at the edge. NXP is one of the latest to do so.
          R-bloggers weekly – top R posts from last week (2018-10-28 till 2018-11-03)      Cache   Translate Page      
Most liked R posts from last week, sorted based on the number of likes they got on twitter, enjoy: From webscraping data to releasing it as an R package to share with the world: a full tutorial (129 likes) How to Create a Correlation Matrix in R (124 likes) Machine Learning Basics – Random Forest (119 likes) Running R Code in Parallel (94 likes) Time series visualizations with wind turbine energy data in R (92 likes) Book Review – Sound Analysis and Synthesis with R (86 likes) Multilevel Modeling Solves the Multiple Comparison Problem: An Example with R (78 likes) Master R shiny: One trick to build maintainable and scalable event chains (75 likes) Communicating results with R Markdown (74 likes) Visualize the Business Value of your Predictive Models with modelplotr (72 likes) R shiny custom docker server with caching (58 likes) Automated Email Reports with R (58 likes) Segmenting Customers by their Purchase Histories Using Non-Negative Matrix Factorization (57 likes) Scatterplot matrices (pair plots) with cdata and ggplot2 (53 likes) Feature Selection with the Boruta Algorithm (52 likes) How to Make a Pie Chart in R (52 likes) New Course: Analyzing Election and Polling Data in R (51 likes) […]
          ‘How do neural nets learn?’ A step by step explanation using the H2O Deep Learning algorithm.      Cache   Translate Page      
In my last blogpost about Random Forests I introduced the codecentric.ai Bootcamp. The next part I published was about Neural Networks and Deep Learning. Every video of our bootcamp will have example code and tasks to promote hands-on learning. While the practical parts of the bootcamp will be using Python, below you will find the English R version of this Neural Nets Practical Example, where I explain how neural nets learn and how the concepts and techniques translate to training neural nets in R with the H2O Deep Learning function. You can find the video on YouTube but as, as before, it is only available in German. Same goes for the slides, which are also currently German only. See the end of this article for the embedded video and slides. Neural Nets and Deep Learning Just like Random Forests, neural nets are a method for machine learning and can be used for supervised, unsupervised and reinforcement learning. The idea behind neural nets has already been developed back in the 1940s as a way to mimic how our human brain learns. That’s way neural nets in machine learning are also called ANNs (Artificial Neural Networks). When we say Deep Learning, we talk about big and complex neural nets, which are able to solve complex tasks, like image or language understanding. Deep Learning has gained traction and success particularly with the recent developments in GPUs and TPUs (Tensor Processing Units), the increase in computing power and data in general, as well as the development of easy-to-use frameworks, like Keras and TensorFlow. We find Deep Learning in our everyday lives, e.g. in voice recognition, computer vision, recommender systems, reinforcement learning and many more. The easiest type of ANN has only node (also called neuron) and is called perceptron. Incoming data flows into this neuron, where a result is calculated, e.g. by summing up all incoming data. Each of the incoming data points is multiplied with a weight; weights can basically be any number and are used to modify the results that are calculated by a neuron: if we change the weight, the result will change also. Optionally, we can add a so called bias to the data points to modify the results even further. But how do neural nets learn? Below, I will show with an example that uses common techniques and principles. Libraries First, we will load all the packages we need: tidyverse for data wrangling and plotting readr for reading in a csv h2o for Deep Learning (h2o.init initializes the cluster) library(tidyverse) library(readr) library(h2o) h2o.init(nthreads = -1) ## Connection successful! ## ## R is connected to the H2O cluster: ## H2O cluster uptime: 3 hours 46 minutes ## H2O cluster timezone: Europe/Berlin ## H2O data parsing timezone: UTC ## H2O cluster version: 3.20.0.8 ## H2O cluster version age: 1 month and 16 days ## H2O cluster name: H2O_started_from_R_shiringlander_jpa775 ## H2O cluster total nodes: 1 ## H2O cluster total memory: 3.16 GB ## H2O cluster total cores: 8 ## H2O cluster allowed cores: 8 ## H2O cluster healthy: TRUE ## H2O Connection ip: localhost ## H2O Connection port: 54321 ## H2O Connection proxy: NA ## H2O Internal Security: FALSE ## H2O API Extensions: XGBoost, Algos, AutoML, Core V3, Core V4 ## R Version: R version 3.5.1 (2018-07-02) Data The dataset used in this example is a customer churn dataset from Kaggle. Each row represents a customer, each column contains customer’s attributes We will load the data from a csv file: telco_data % select_if(is.numeric) %__% gather() %__% ggplot(aes(x = value)) + facet_wrap(~ key, scales = "free", ncol = 4) + geom_density() ## Warning: Removed 11 rows containing non-finite values (stat_density). … and barcharts for categorical variables. telco_data %__% select_if(is.character) %__% select(-customerID) %__% gather() %__% ggplot(aes(x = value)) + facet_wrap(~ key, scales = "free", ncol = 3) + geom_bar() Before we can work with h2o, we need to convert our data into an h2o frame object. Note, that I am also converting character columns to categorical columns, otherwise h2o will ignore them. Moreover, we will need our response variable to be in categorical format in order to perform classification on this data. hf % mutate_if(is.character, as.factor) %__% as.h2o Next, I’ll create a vector of the feature names I want to use for modeling (I am leaving out the customer ID because it doesn’t add useful information about customer churn). hf_X
          Technical Architect - Data Solutions - CDW - Milwaukee, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Milwaukee, WI jobs
          Technical Architect - Data Solutions - CDW - Madison, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Madison, WI jobs
          Director, Marketing Automation for Relationship Marketing - Microsoft - Redmond, WA      Cache   Translate Page      
In addition, this role will lead the identification and leveraging of the latest marketing capabilities like machine learning and AI to propel Relationship...
From Microsoft - Wed, 31 Oct 2018 07:57:36 GMT - View all Redmond, WA jobs
          Three ways to avoid bias in machine learning      Cache   Translate Page      
At this moment in history it’s impossible not to see the problems that arise from human bias. Now magnify that by compute and you start to get a sense for just how dangerous human bias via machine learning can be.
          AI δυνατότητες στο Galaxy S10 λόγω dual-core AI chip;      Cache   Translate Page      
Ένας από τους τομείς όπου το Galaxy S10 θα φέρει βελτιώσεις, σύμφωνα με πληροφορίες, είναι η τεχνητή νοημοσύνη μιας και θα διαθέτει ξεχωριστό chip για αυτές τις λειτουργίες, ακολουθώντας την τακτική άλλων κατασκευαστών smartphone. Πιο συγκεκριμένα, οι πληροφορίες αναφέρουν ότι η Samsung θα εφοδιάσει το Galaxy S10 με ξεχωριστό NPU Chip (Neural Processing Unit), το οποίο θα μπορεί να διαχειρίζεται πιο πολύπλοκες εφαρμογές τεχνητής νοημοσύνης και machine learning. Επιπλέον, ο επεξεργαστής για την τεχνητή νοημοσύνη θα είναι dual-core σαν αυτόν που βρίσκεται στον Kirin 980 της Huawei.
Αυτό που έχει κάνει η Huawei είναι να εφοδιάσει τον επεξεργαστή της με διπλή μονάδα επεξεργασίας νευρωνικών δικτύων τεχνητής νοημοσύνης (Dual NPU), πετυχαίνοντας την αναβάθμιση της εμπειρίας της AI με μεγαλύτερη επεξεργαστική ισχύ, ανάλογο με αυτό που αναμένεται να πετύχει και η Samsung εφόσον λειτουργήσει κατά τον ίδιο τρόπο, ενώ ταυτόχρονα θα πετύχει μείωση στην κατανάλωση ενέργειας, κάτι το οποίο θα πρέπει να περιμένουμε να το δούμε στην πράξη. Αν όμως λάβουμε υπ’ όψιν το ότι ο Exynos επεξεργαστής που θα βρίσκεται στα Galaxy S10 και S10 Plus θα βασίζεται στην αρχιτεκτονική των 7nm LPP (Low Power Plus), τότε η κατανάλωση ενέργειας αναμένεται να είναι ακόμα χαμηλότερη, με τη Samsung να είναι η πρώτη κατασκευάστρια εταιρεία που θα χρησιμοποιήσει αυτήν την τεχνική.
          Chief Architect (Data Analytics - Strategic Technology Centre) - Singapore Technologies Engineering Ltd - Ang Mo Kio      Cache   Translate Page      
Location based services using Google Map, ESRI, Trillium etc. is a plus. ST Engg has identified data analytics, machine learning and Artificial Intelligence as...
From Singapore Technologies Electronics - Fri, 12 Oct 2018 06:27:13 GMT - View all Ang Mo Kio jobs
          Model & Controls Assurance Analyst - Intuit - Mountain View, CA      Cache   Translate Page      
Experience with Statistical and Machine learning Classification and Regression techniques. Help build the framework for establishing a governance function to...
From Intuit - Tue, 16 Oct 2018 05:15:18 GMT - View all Mountain View, CA jobs
          Sr. Software Development Engineer - Amazon.com - Bellevue, WA      Cache   Translate Page      
Build Products for amazon external facing and internal facing systems. The team uses various content classification and machine learning algorithms for solving...
From Amazon.com - Mon, 05 Nov 2018 19:19:34 GMT - View all Bellevue, WA jobs
          Software development manager - Amazon.com - Bellevue, WA      Cache   Translate Page      
Build Products for amazon external facing and internal facing systems. The team uses various content classification and machine learning algorithms for solving...
From Amazon.com - Wed, 18 Jul 2018 19:20:37 GMT - View all Bellevue, WA jobs
          Senior Solutions Architect, Alexa - Amazon.com - Seattle, WA      Cache   Translate Page      
5+ years experience with machine learning systems and model training. Amazon is an Equal Opportunity-Affirmative Action Employer – Minority / Women / Disability...
From Amazon.com - Fri, 02 Nov 2018 07:19:20 GMT - View all Seattle, WA jobs
          Lifecycle Marketing Manager - Amazon.com - Seattle, WA      Cache   Translate Page      
Business Skills Needed:. You will work closely with our Business Intelligence and Machine Learning teams to evolve our understanding of key predictors of...
From Amazon.com - Fri, 02 Nov 2018 01:19:22 GMT - View all Seattle, WA jobs
          Business Intelligence Engineer - II - Amazon.com - Seattle, WA      Cache   Translate Page      
Experience with forecasting measures and machine learning. What does a healthy organization look like in Amazon?...
From Amazon.com - Thu, 01 Nov 2018 19:20:12 GMT - View all Seattle, WA jobs
          Business Analyst - I - Amazon.com - Seattle, WA      Cache   Translate Page      
Experience with forecasting measures and machine learning. What does a healthy organization look like in Amazon?...
From Amazon.com - Thu, 01 Nov 2018 19:20:06 GMT - View all Seattle, WA jobs
          Imaging and Machine Learning Research Scientist - AIRY:3D - Montréal, QC      Cache   Translate Page      
Contribute to expanding our patent portfolio, generate publishable research. We’re looking to add even more imaging and machine learning expertise to our team....
From Indeed - Wed, 17 Oct 2018 15:54:14 GMT - View all Montréal, QC jobs
          Machine learning Python hacks, creepy Linux commands, Thelio, Podman, and more      Cache   Translate Page      
I'm filling in again for this weeks top 10 while  Rikki Endsley  is recovering from  LISA 18  held last week in Nashville, Tennessee. We're starting to gather articles for our 4 th annual Open Source ... - Source: opensource.com
          [TobagoJack] Per imperatives => solutions Danger = opportunity The sort of equations some fi...      Cache   Translate Page      
Per imperatives => solutions
Danger = opportunity
The sort of equations some find challenging to take on board or even to understand, due to arms and legs mentality and beans counting

... let us see if brave-new-world protocol goes right, and if so, exports ala free-trades

Am wondering if Africans should say “no” to bot-doctors as part of turning head away from hospitals and railroads :0)

Am also wondering whether the bots would be practicing wholistic eastern medicine or specific-toxic western healing, and

What happens when bots prescribe in accordance w/ traditional Napalese healing but skip licensing fee.

Also, would such medicine improve UK healthcare system

brinknews.com

China’s Doctor Shortage Can Be Solved by AI
Andy HoNovember 6, 2018

A surgeon performs an operation at a clinic in the southwest Chinese city of Chongqing. AI might be able to solve China's doctor shortage problem.

Photo: Peter Parks/AFP/Getty Images

If there is one country that has invested heavily in health care reform over the last few years, it is China. But as its population grows older, with already 300 million people suffering from chronic diseases, it seems almost impossible to keep up with the soaring demand for health care. According to the latest data from the Organisation for Economic Co-operation and Development, China has 1.8 practicing doctors per 1,000 citizens, compared to 2.6 for the U.S. and 4.3 for Sweden. Can artificial intelligence relieve China’s overworked doctors of some of their burdens?

China’s Ailing Health Care SystemThe hard-working medical professionals who keep China’s ailing health care system running could certainly use a helping hand. Overcrowding is the order of the day in the country’s urban hospitals, with a typical outpatient department in Beijing seeing about 10,000 people every day. The problem is exacerbated by the scarcity of medical facilities in rural areas, which causes people to flock to hospitals in nearby cities.

As the Future Health Index 2018 by Philips shows, the relatively low number of skilled health care professionals in relation to the size of the population is one of the main reasons why access to care in China lags behind most of the other fifteen countries surveyed.

Demographic projections give further reason for concern. The demand for care will only continue to grow as China is aging more rapidly than almost any country in the world. The United Nations estimates that by 2040, the country’s population over 65 will reach about 303 million, which is almost equal to the current total population of the U.S.

However, there is also reason for optimism.

In its commitment to offer accessible and affordable care for all, the Chinese government is spearheading the development of health care technologies. And perhaps the most promising is AI.



The Rise of AIAI can help make sense of large amounts of data, fueled by computing power that has risen dramatically over the last few years. That’s why China offers particularly fertile ground for AI development: With its 1.4 billion population, the country sits on massive troves of data.

Recognizing the country’s AI potential, the government has set out an ambitious plan to turn China into the world’s leading AI innovation center. Health care is one of the industries that are set to benefit from multibillion-dollar investments in startups, academic research, and moonshot projects. This is not merely a vision, but a reality already in the making. According to Yiou Intelligence, a Beijing-based consultancy firm, some 131 Chinese companies are currently working on applying AI in health care.



A Smart Personal Assistant for PhysiciansSpeeding up the screening of medical images is just one of the ways in which AI could relieve China’s overburdened health care system.

As one Chinese radiologist said in an interview with The New York Times: “We have to deal with a vast amount of medical images every day. So we welcome technology if it can relieve the pressure while boosting efficiency and accuracy.”

We should take these needs to heart and focus on developing intelligent applications that ease the workload for physicians while improving outcomes for patients. Crucially, the goal should not be to replace physicians, but to augment their impact in their daily work, strengthening their role in the delivery of efficient and high-quality care.

For some, AI conjures up images of autonomous robots replacing human workers. But I believe that in health care, AI is best thought of as a smart personal assistant for physicians that adapts to their needs and ways of working—“adaptive intelligence,” as we call it at Philips. Viewed through that lens, AI will make health care more—not less—human.

Today, AI is already helping physicians with the analysis of medical images. As AI becomes increasingly sophisticated and is integrated with medical knowledge, it could support ever more precise diagnoses and personalized treatment plans. But in the short term, arguably the greatest gains are to be made in solving operational bottlenecks in hospitals—for example, by helping physicians get a quick overview of all clinically relevant information on a patient.

Patient data are usually stored in many disparate systems and formats. At Zhongshan Hospital in Shanghai, it can take a physician up to 20 days to manually extract all relevant information from 200 unstructured medical reports into one structured format.

By combining AI methods like natural language processing and machine learning with clinical knowledge, it is possible to collate all clinically relevant information in one dashboard. Physicians could spend less time capturing information from unstructured reports and less time sitting in front of a screen to get a complete picture of the patient.

Improving Care Close to People’s HomesAI could also enable patients with chronic conditions to become more informed about their health and to stay connected with professional caregivers.

According to the Future Health Index 2018, adoption of telehealth in China is currently much lower than the 16-country average, but the Chinese population is open to the use of technologies that can supplement the care they currently receive.

For example, home health monitoring technology powered by AI could help the frail and elderly stay connected with professional caregivers to ensure they receive timely care when needed. People with diabetes or hypertension could benefit from similar technology that allows them to track their condition via clinically validated sensors and devices.

Such initiatives would fit perfectly with the Chinese government’s ambition to improve care at the grassroots level to counter congestion in city hospitals. More widespread adoption of AI technologies should go hand in hand with investments in primary care facilities and Internet connectivity in rural areas—making health care more equally accessible and affordable and allowing people to enjoy a better quality of life close to their communities.

Looking further ahead, AI could also become pivotal in addressing lifestyle-related diseases such as obesity—a major health concern that affects about one in eight people in China. Imagine people with high risk of obesity getting bespoke lifestyle tips via their smartphone. On a population level, data analyses could inform public interventions targeted at specific age groups or geographic areas. As the Chinese government has outlined in its plans for a “Healthy China 2030,” the focus of the health care system will increasingly shift from treatment to prevention.

A Call for CollaborationHow to accelerate this journey toward more efficient, accessible and preventative care?

First, building a more robust data ecosystem should be top priority. The quality of AI is only as good as the quality of the data fed into it. China’s health care system would benefit from shared data standards, interoperability of systems, and improved data exchange protected by top-notch security measures. The establishment of three national digital databases with health information by 2020 is an important step in this direction.

Second, data-driven approaches such as AI will only have the desired impact when combined with proven medical expertise. AI is only part of any solution; it is never a solution by itself. A deep understanding of the clinical context is indispensable. Any form of AI-assisted care must be centered on the physician and the patient, taking their needs as a starting point and building on the wealth of human knowledge that is already available.

Third, AI-enabled tools must be rigorously tested against the highest regulatory standards. In health care, where lives are at stake, we need to deploy new technologies wisely and carefully. Only with proper clinical validation can we ensure responsible, safe and effective use of AI. Physicians as well as patients also require education on a tool’s strengths and limitations.

Fourth, collaboration between academia, startups, and established companies is of paramount importance. The challenges in China’s health care system are simply too big for any player to address it alone. In this light, it is encouraging that the Chinese government has recently founded a collaborative platform to promote the exchange of ideas and kick-start new projects in intelligent medicine.

Finally, to ensure we are creating a future-ready health care system in China, we must address the shortage of talent at the intersection of medicine and data science. We should nurture and invest in developing people who combine medical know-how with a firm understanding of AI and other technologies. Ultimately, the sustainability of China’s health care system may lie in their hands.

This piece first appeared on the World Economic Forum Agenda.

          Your Client Engagement Program Isn't Doing What You Think It Is.      Cache   Translate Page      

Amazing products without engaged clients are bound to fail, and companies claiming to have found the single best solution to client engagement are only fooling themselves.

What seems to work today to keep your clients engaged won't necessarily work tomorrow. The "optimal" client engagement tactic for your product will change over time and companies must be fluid and adaptable to accommodate ever-changing client needs and business strategies. Becoming complacent by settling for a strategy that works "for now" or "well enough" leads to risk aversion and unrealized potential. Constant recalibration is crucial, yet exploration can be costly and may lead nowhere. A principled approach to finding the right client engagement tactics at any point in time is essential.

Enter, Bandits!

Here at Stitch Fix, we prioritize personalization in every communication, interaction, and outreach opportunity we have with clients. Contextual bandits are one of the ways we enable this personalization.

In a nutshell, a contextual bandit is a framework that allows you to use algorithms to learn the most effective strategy for each individual client, while simultaneously using randomization to continuously track how successful each of your different action choices are.

Implementing a contextual-bandit-based client engagement program will allow you to:

  1. Understand how the performance of your tactics change over time;
  2. Select a personalized tactic for each client based on his or her unique characteristics;
  3. Introduce new tactics relevant to subpopulations of clients in a systematic manner; and
  4. Continuously refine and improve your algorithms.

Part 1: There are Significant Limitations to Typical Client Engagement Approaches.

Let's set up a simple, clear example.

Current state: There are a group of individuals who have used your product in the past, but are no longer actively engaged. You want to remind them about your great product.

Proposed idea: We'll do an email campaign. Clients who haven't interacted with you for a while will be eligible to receive this email, and the purpose is to get them to visit your website. (Note that instead of email, we could just as easily use a widget on a website, a letter in the mail, or any other method of communicating with clients.)

The One Size Fits All Method

Your team brainstorms several tactics and decides to run a test to see which one works best. Let's use a simple three-tactic example here:

  1. Tactic A: no email. This is our control, which we can use to establish a baseline for client behavior;
  2. Tactic B: an emailed invitation to work 1-on-1 with (in our case) an expert stylist via email to make sure we get you exactly what you want; and
  3. Tactic C: an emailed promotional offer.

You run this test and find that Tactic B works the best. As such, you decide to scale Tactic B out to all clients, since it is expected to maximize return.

This is an endpoint in many marketing and product pipelines. The team celebrates the discovery of the "best" strategy and now it's time to move on to the next project; Case closed.

Main Limitations with this Approach

  1. You have no idea how long Tactic B will continue to be the best. Let's say over the next year your product improves and expands, leading to a change in the demographics of your client base. How confident are you that Tactic B is still the best?
  2. There is no personalization. All clients are receiving Tactic B. Some subsegment of clients likely would have performed better with Tactic C, but by scaling out Tactic B to all clients, these clients did not receive their optimal tactic.
  3. You are not taking advantage of key pieces of information. Since everyone in this audience was a previous client, you have information on how they interacted with you in the past! To address this, teams frequently build out a decision tree that groups their clients into broad categories (such as by age or tenure with the company). However…
  4. Using decision trees to group clients into categories leads to unoptimized outreach programs. Each category of clients might get a different tactic, and as new tactics or categories are created, these trees can grow larger and larger. Many of you have seen it: 10+ branch decision trees that try to segment a client base into different categories. When these trees grow too large, not only do they become difficult to manage, but it becomes more and more questionable whether or not you are actually doing what you think you are doing.
Does this look like your program?

Part 2: Multi-Armed Bandits Allows Continuous Monitoring

The above situation is common, suboptimal, and headache-inducing. Developing a testing and implementation strategy with bandits can remove or reduce many of these limitations.

A standard multi-armed bandit is the most basic bandit implementation. It allows us to allocate a small amount of clients to continuously explore how different tactics are performing, while giving the majority of clients the current one-size-fits-all best-performing tactic. The standard implementation updates which tactic is best after every client interaction, allowing you to quickly settle on the most effective large-scale tactic.

To get into more technical detail, a multi-armed bandit is a system where you must select one action from a set of possible actions for a given 'resource' (in our example, the 'resource' is a client). The 'reward' (if a client responds to the offer or not) for the selected action is exposed, but the reward for all other actions remains unknown (we don't know what a specific client would have done if we had sent them a different offer!).

The reward for a given action can be thought of as a random reward drawn from a probability distribution specific to that action. Because these probability distributions may not be known or may change over time, we want this system to allocate some resources to improving our understanding of the different choices ('exploration') while simultaneously maximizing our expected gains based on our historical knowledge ('exploitation'). Our goal is to minimize "regret," defined as the difference between the sum of rewards if we used an optimal strategy and the actual sum of rewards realized.

Mathematically, if we have a bandit with K choices (K arms), we can define a reward for 1 ≤ i ≤ K and n ≥ 1, where i is an arm of the bandit and n is the round we are currently considering.

Each round yields a reward The reward of each round is a function of which arm was selected for the client, and is assumed to be independent of both the rewards from previous rounds and the distributions of the other arms, and only dependent upon the probability distribution associated with the selected arm.

We thus want to minimized our regret, defined as:

, where total number of rounds, maximum reward, and is the reward in round n from selecting arm i.

There are numerous strategies that can be utilized to select how clients are allocated to either exploitation or exploration in order to minimize the regret of your bandit. For this article, we will consider the simplest, called the epsilon-greedy algorithm. In the epsilon-greedy algorithm, the best action is selected for 1−ϵ of your audience entering your program, and a random action is selected for the remaining ϵ of your audience. ϵ can be set to any value, depending on how many resources you want to allocate to exploration. For example, if ϵ is set to 0.1, then 10% of your audience is being directed to exploration, and 90% of your audience is being directed to your best tactic. If desired, ϵ can be decreased over time to reduce the total regret of your system. Other popular allocation strategies that can reduce the overall regret of your system include Thompson sampling and UCB1[1][2].

Back to our Example

Let’s get back to our 3-tactic email test, and set up a standard multi-armed bandit. We need to decide to reach out to a client with either Tactic A, B, or C. We are going to use an epsilon-greedy approach and set ϵ=0.1. 10% of our clients will randomly be presented with one of the three tactics. The other 90% will receive whichever tactic is the current top-performer. Each time we give a client a tactic, we update the score for that tactic. We then examine the performance of the different tactics, and re-assign the best-performing tactic to the one with the highest score.

In the early stages of a standard multi-armed bandit, switching may occur frequently in the first few passes, but the pipeline will quickly stabalize.

The above animation demonstrates a simple, standard multi-armed bandit to help get you started thinking about how a bandit implementation might look. This example already allows us to continuously monitor how tactics are performing and actively redirects most of our audience to the current "best" tactic, but there is plenty more we can do to improve performance! For example, if we want to further minimize regret, we would want to use a more powerful regret-minimization technique (UCB1, Thompson Sampling, etc.) If we are nervous about our client population changing over time, we can implement a forgetting factor to down-weight older data points. New tactics can be added to this framework in a similar manner.

Part 3: Personalize Outreach with Contextual Bandits

While we have already improved upon our original example, let's take this a step further to get to true personalization.

Contextual bandits provide an extension to the bandit framework where a context (feature-vector) is associated with each client.[3] This allows us to personalize the action taken on each individual client, rather than simply applying the overall best tactic. For example, while Tactic B might perform the best if applied to all clients, there are certainly some clients who would respond better to Tactic C.

What this means in terms of our example is that instead of 90% of our clients being sent the one-size-fits-all best email, these clients instead enter our "selection algorithm." We then read in relevant features of these clients to decide which outreach tactic best suits each client’s needs and results in the best outcomes. We continue to assign 10% clients to a random tactic, because this lets us know the unbiased performance of each tactic and provides us with data to periodically retrain our selection algorithm.

To get started, we need to take some unbiased data and train a machine learning model. Let’s assume we have been running the multi-arm bandit above: because we randomly assigned clients to our different tactics, we can use this data to train our algorithm! The point here is to understand which clients we should be assigning to each tactic, or, for a given client, which tactic gives this client the highest probability of having a positive outcome.

In a Contextual Bandit, we use client features to select the best tactic for the majority of our clients and continue to pull in data from our randomly allocated clients to retrain our algorithm.

While a multi-armed bandit can be relatively quick to implement, a contextual bandit adds a decent amount of complexity to our problem. In addition to training and applying a machine learning model to the majority of clients, there are a number of additional behind-the-scenes steps that must be taken: for example, we need to establish a cadence for retraining this model as we continue to acquire more data, as well as build out significantly more logging in order to track from which pathway clients are being assigned to different tactics.

Why do we care about which pathway clients received an offer from? If we use an algorithm to assign a tactic to a client, clients with certain characteristics are more likely to be assigned to certain tactics. This is why it is important to have some clients assigned randomly: we can be confident that, in the random assignments, the underlying client distributions are the same for all tactics.

Contextual Considerations

Depending on the regret-minimization strategy, there may be restrictions on the type of model you can build. For example, using an epsilon-greedy strategy allows you to use any model you like. However, with Thompson sampling, using a non-linear model introduces significant additional challenges.[5] When deciding how to approach regret minimization, make sure to take your desired modeling approach into account.

No implementation is perfect, and bandits are no exception. Both multi-armed bandits and contextual bandits are best utilized when we have a clear, well-defined 'reward' – maybe this is clicking on a webpage banner, or clicking through an email, or purchasing something. If you can't concretely define a reward, you can't say whether or not a tactic was successful, let alone train a model. In addition, if you have significantly delayed feedback you will have to do some additional work to get everything running smoothly[4]. Contextual bandits also lead to more complex code, which gives more room for things to break when compared to implementing simpler strategies.

Finally, think carefully about real-world restrictions: for example, while a contextual bandit can support a large number of tactics, if you have very few clients entering the program you are not going to be gaining much information from your random test group ϵ . In this case, it may be wise to restrict the number of tactics to something manageable. Remember though, you can always remove tactics to make room for new ones!

Part 4: Summary and Final Thoughts

Expanding your program from using business-logic-driven decision making to model-based decision making can significantly improve the performance of your client engagement strategies, and bandits can be a great tool to facilitate this transition.

While the example we used involved email, this same pipeline can be applied in a variety of different domains. Other examples include prioritizing widgets on a webpage, modifying the flow clients experience as they click through forms on your website, or many other situations where you are performing client outreach.

If you currently have a single tactic scaled out to all clients and are not gathering any unbiased data, one viable approach to improve this situation would be to first transition to a multi-armed bandit, and then at a later timepoint transition to a contextual bandit. A basic multi-armed bandit can be quick to implement and allow you to begin gathering unbiased data. Eventually, utilizing a contextual bandit in your client engagement strategy will allow you to active adjust to changing client climates and needs, continuously test strategies, and personalize, personalize, personalize!

References

[1]↩ Minimizing Regret: https://link.springer.com/content/pdf/10.1023%2FA%3A1013689704352.pdf
[2]↩ Thompson sampling: http://proceedings.mlr.press/v23/agrawal12/agrawal12.pdf
[3]↩ Contextual bandits seminal paper: https://arxiv.org/pdf/1003.0146.pdf
[4]↩ Delayed Feedback: https://arxiv.org/pdf/1803.10937.pdf
[5]↩ Another Thompson Sampling Paper: https://pdfs.semanticscholar.org/1c21/2b33a91d7b1c9878af0395d4992a6d4e0d54.pdf

          Critical components for implementing AI      Cache   Translate Page      
Douglas Reding, MD, chief medical officer for Ascension Wisconsin and practicing oncologist, talks about Ascension's journey with its precision medicine task force, the need for AI support and the future of AI and machine learning.
          Machine Learning Engineer / Algorithm Developer - TECHNICA CORPORATION - Dulles, VA      Cache   Translate Page      
Job Description: We are seeking a highly creative software engineer experienced in artificial intelligence and deep learning techniques to design, develop,...
From Technica Corporation - Fri, 05 Oct 2018 10:31:19 GMT - View all Dulles, VA jobs
          Machine Learning Engineer - TECHNICA CORPORATION - Dulles, VA      Cache   Translate Page      
Technica Corporation is seeking a Machine Learning. Engineer to support our internal Innovation, Research....
From Technica Corporation - Thu, 23 Aug 2018 10:27:08 GMT - View all Dulles, VA jobs
          Open Source Machine Learning Tool Could Help Choose Cancer Drugs      Cache   Translate Page      
November 6, 2018 Atlanta, GA A new decision support tool could help clinicians choose the right chemotherapy drugs. Research Biotechnology, Health, Bioengineering, Genetics Cancer Research Life Sciences and Biology
          NinePoint Keeps AI Offerings Fresh with IRIS      Cache   Translate Page      

FDA has just given a nod to NinePoint Medical’s Intelligent Real-time Image Segmentation (IRIS) software upgrade for the NvisionVLE Imaging System. The Bedford, MA-based company said IRIS is an artificial intelligence-based platform for image feature segmentation is the first tool of its kind approved for use in imaging of esophageal tissue.

The NvisionVLE Imaging System allows physicians to perform a Volumetric Laser Endomicroscopy (VLE) procedure that produces real-time, high-resolution cross-sectional images. This system enables gastroenterologists to thoroughly evaluate the esophageal tissue surface and sub-surface for potential disease that may not be visible with conventional medical imaging technologies such as endoscopy and ultrasound.

The new IRIS upgrade uses machine learning algorithms to segment and visualize esophageal image features in real-time, to assist clinicians as they identify and target regions of interest during an endoscopic procedure.

“The clearance of the IRIS product marks the successful culmination of a multi-year development effort within our organization, and with [FDA],” Eman Namati, PhD, president and CEO of NinePoint Medical said in a release. “We are eager to roll this out in a controlled-market release in the coming months prior to making the technology more broadly available.”

AI has been a hot topic in medtech in 2018 and MD+DI has been on the frontlines covering the recent AI boom. That coverage spawned a webinar titled, How Artificial Intelligence Has Changed Everything for Medtech hosted by Dave Saunders CTO and Co-founder of Galen Robotics. In addition, MD+DI listed six companies with AI-based technology slated to rock the medtech sector. 


          Medicare Fraud Detection Becoming Possible Through Machine-Learning Algorithms      Cache   Translate Page      

Researchers from Florida Atlantic University’s College of Engineering and Computer Science published a study in Health Information Science and Systems that shows how machine learning and advanced analytics could lead to Medicare fraud detection. The breakthrough could lead to $19-65 billion annual savings of Medicare funds lost to fraud.

The researchers tested six different machine learners on both balanced and imbalanced data sets using Medicare Part B data, ultimately finding the RF100 random forest algorithm to be the most effective in detecting potential fraudulent claims, and that imbalanced data sets provided the most accurate results.

Click here to continue and read more...


          Protecting What Matters: Defining Data Guardrails and Behavioral Analytics      Cache   Translate Page      

Posted under: General

This is the second post in our series on Protecting What Matters: Introducing Data Guardrails and Behavioral Analytics. Our first post, Introducing Data Guardrails and Behavioral Analytics: Understand the Mission, introduced the concepts and outlined the major categories of insider risk. This post defines the concepts.

Data security has long been the most challenging domain of information security, despite being the centerpiece of our entire practice. We only call it “data security” because “information security” was already taken. Data security must not impede use of the data itself. By contrast it’s easy to protect archival data (encrypt it and lock the keys up in a safe). But protecting unstructured data in active use by our organizations? Not so easy. That’s why we started this research by focusing on insider risks, including external attackers leveraging insider access. Recognizing someone performing an authorized action, but with malicious intent, is a nuance lost on most security tools.

How Data Guardrails and Data Behavioral Analytics are Different

Both data guardrails and data behavioral analytics strive to improve data security by combining content knowledge (classification) with context and usage. Data guardrails leverage this knowledge in deterministic models and processes to minimize the friction of security while still improving defenses. For example, if a user attempts to make a file in a sensitive repository public, a guardrail could require them to record a justification and then send a notification to Security to approve the request. Guardrails are rule sets that keep users “within the lines” of authorized activity, based on what they are doing.

Data behavioral analytics extends the analysis to include current and historical activity, and uses tools such as artificial intelligence/machine learning and social graphs to identify unusual patterns which bypass other data security controls. Analytics reduces these gaps by looking not only at content and simple context (as DLP might), but also adding in history of how that data, and data like it, has been used within the current context. A simple example is a user accessing an unusual volume of data in a short period, which could indicate malicious intent or a compromised account. A more complicated situation would identify sensitive intellectual property on an accounting team device, even though they do not need to collaborate with the engineering team. This higher order decision making requires an understanding of data usage and connections within your environment.

Central to these concepts is the reality of distributed data actively used widely by many employees. Security can’t effectively lock everything down with strict rules covering every use case without fundamentally breaking business processes. But with integrated views of data and its intersection with users, we can build data guardrails and informed data behavioral analytical models, to identify and reduce misuse without negatively impacting legitimate activity. Data guardrails enforce predictable rules aligned with authorized business processes, while data behavioral analytics look for edge cases and less predictable anomalies.

How Data Guardrails and Data Behavioral Analytics Work

The easiest way to understand the difference between data guardrails and data behavioral analytics is that guardrails rely on pre-built deterministic rules (which can be as simple as “if this then that”), while analytics rely on AI, machine learning, and other heuristic technologies which look at patterns and deviations.

To be effective both rely on the following foundational capabilities:

  • A centralized view of data. Both approaches assume a broad understanding of data and usage – without a central view you can’t build the rules or models.
  • Access to data context. Context includes multiple characteristics including location, size, data type (if available), tags, who has access, who created the data, and all available metadata.
  • Access to user context, including privileges (entitlements), groups, roles, business unit, etc.
  • The ability to monitor activity and enforce rules. Guardrails, by nature, are preventative controls which require enforcement capabilities. Data behavioral analytics can be used only for detection, but are far more effective at preventing data loss if they can block actions.

The two technologies then work differently while reinforcing each other:

  • Data guardrails are sets of rules which look for specific deviations from policy, then take action to restore compliance. To expand our earlier example:
    • A user shares a file located in cloud storage publicly. Let’s assume the user has the proper privileges to make files public. The file is in a cloud service so we also assume centralized monitoring/visibility, as well as the capability to enforce rules on that file.
    • The file is located in an engineering team’s repository (directory) for new plans and projects. Even without tagging, this location alone indicates a potentially sensitive file.
    • The system sees the request to make the file public, but because of the context (location or tag), it prompts the user to enter a justification to allow the action, which gets logged for the security team to review. Alternatively, the guardrail could require approval from a manager before allowing the action.

Guardrails are not blockers because the user can still share the file. Prompting for user justification both prevents mistakes and loops in security review for accountability, allowing the business to move fast while minimizing risk. You could also look for large file movements based on pre-determined thresholds. A guardrail would only kick in if the policy thresholds are violated, and then use enforcement actions aligned with business processes (such as approvals and notifications) rather than simply blocking activity and calling in the security goons.

  • Data behavioral analytics use historical information and activity (typically with training sets of known-good and known-bad activity), which produce artificial intelligence models to identify anomalies. We don’t want to be too narrow in our description, because there are a wide variety of approaches to building models.
    • Historical activity, ongoing monitoring, and ongoing modeling are all essential – no matter the mathematical details.
    • By definition we focus on the behavior of data as the core of these models, rather than user activity; this represents a subtle but critical distinction from User Behavioral Analytics (UBA). UBA tracks activity on a per-user basis. Data behavioral analytics (the acronym DBA is already taken, so we’ll skip making up a new TLA), instead looks at activity at the source of the data. How has that data been used? By which user populations? What types of activity happen using the data? When? We don’t ignore user activity, but we track usage of data.
      • For example we could ask, “Has a file of this type ever been made public by a user in this group?” UBA would ask “Has this particular user ever made a file public?” Focusing on the data offers a chance potential to catch a broader range of data usage anomalies.
    • At risk fo stating the obvious, the better the data, the better the model. As with most security-related data science, don’t assume more data inevitably produces better models. It’s about the quality of the data. For example social graphs of communication patterns among users could be a valuable feed to detect situations like files moving between teams who do not usually collaborate. That’s worth a look, even if you wouldn’t want to block the activity outright.

Data guardrails handle known risks, and are especially effective at reducing user error and identifying account abuse resulting from tricking authorized users into unauthorized actions. Guardrails may even help reduce account takeovers, because attackers cannot misuse data if their action violate a guardrail. Data behavioral analytics then supplements guardrails for unpredictable situations and those where a bad actor tries to circumvent guardrails, including malicious misuse and account takeovers.

Now you have a better understanding of the requirements and capabilities of data guardrails and data behavioral analytics. Our next post will focus on some quick wins to justify including these capabilities in your data security strategy.

- Rich (0) Comments Subscribe to our daily email digest
          “Statistical and Machine Learning forecasting methods: Concerns and ways forward”      Cache   Translate Page      

Roy Mendelssohn points us to this paper by Spyros Makridakis, Evangelos Spiliotis, and Vassilios Assimakopoulos, which begins: Machine Learning (ML) methods have been proposed in the academic literature as alternatives to statistical ones for time series forecasting. Yet, scant evidence is available about their relative performance in terms of accuracy and computational requirements. The purpose […]

The post “Statistical and Machine Learning forecasting methods: Concerns and ways forward” appeared first on Statistical Modeling, Causal Inference, and Social Science.


          Machine Learning Engineer - Workbridge Associates - Toronto, ON      Cache   Translate Page      
One of my best clients is hiring for Machine Learning Engineers and are willing to pay the big bucks for the right talent.... $120 - $155 a day
From Workbridge Associates - Mon, 05 Nov 2018 14:26:59 GMT - View all Toronto, ON jobs
          Micron's bet: Quad-level cell NAND SSDs will finally replace HDDs      Cache   Translate Page      
The company's Micron 5210 ION enterprise SATA SSD is now generally available and aimed at artificial intelligence, machine learning, deep learning and other intensive workloads.
          AI startup Flex Logix touts vastly higher performance than Nvidia      Cache   Translate Page      
Four-year-old startup Flex Logix has taken the wraps off its novel chip design for machine learning. CEO Geoff Tate describes how the chip may take advantage of an "explosion" of inferencing activity in "edge computing," and how Nvidia can't compete on performance.
          Chief Architect (Data Analytics - Strategic Technology Centre) - Singapore Technologies Engineering Ltd - Ang Mo Kio      Cache   Translate Page      
Location based services using Google Map, ESRI, Trillium etc. is a plus. ST Engg has identified data analytics, machine learning and Artificial Intelligence as...
From Singapore Technologies Electronics - Fri, 12 Oct 2018 06:27:13 GMT - View all Ang Mo Kio jobs
          Machine learning Python hacks, creepy Linux commands, Thelio, Podman, and more      Cache   Translate Page      
I[he]#039[/he]m filling in again for this weeks top 10 while Rikki Endsley is recovering from LISA 18 held last week in Nashville, Tennessee. We[he]#039[/he]re starting to gather articles for our 4th annual Open Source Yearbook, get your proposals in soon. Enjoy this weeks[he]#039[/he] top 10.
          11-28-2018 Decentralized Signal Processing and Distributed Control for Collaborative Autonomous Sensor Networks       Cache   Translate Page      
Speaker: Ryan Alan Goldhahn & Priyadip Ray, Lawrence Livermore National Laboratory Talk Title: Decentralized Signal Processing and Distributed Control for Collaborative Autonomous Sensor Networks Series: Center for Cyber-Physical Systems and Internet of Things Abstract: Collaborative autonomous sensor networks have recently been used in many applications including inspection, law enforcement, search and rescue, and national security. They offer scalable, low-cost solutions which are robust to the loss of multiple sensors in hostile or dangerous environments. While often comprised of less capable sensors, the performance of a large network can approach the performance of far more capable and expensive platforms if nodes are effectively coordinating their sensing actions and data processing. This talk will summarize work to date at LLNL on distributed signal processing and decentralized optimization algorithms for collaborative autonomous sensor networks, focusing on ADMM-based solutions for detection/estimation problems and sequential and/or greedy optimization solutions which maximize submodular functions such as mutual information. Biography: Ryan Goldhahn holds a Ph.D. in electrical engineering from Duke University with a focus in statistical and model-based signal processing. Ryan joined the NATO Centre for Maritime Research and Experimentation (CMRE) as a researcher in 2010 and later as the project lead for an effort to use multiple unmanned underwater vehicles (UUVs) to detect and track submarines using multi-static active sonar. In this work he developed collaborative autonomous behaviors to optimally reposition UUVs to improve tracking performance without human intervention. He led several experiments at sea with submarines from multiple NATO nations. At LLNL Ryan has continued to work and lead projects in collaborative autonomy and model-based and statistical signal processing in various applications. He has specifically focused on decentralized detection/estimation/tracking and optimization algorithms for autonomous sensor networks. Priyadip Ray received a Ph.D. degree in electrical engineering from Syracuse University in 2009. His Ph.D. dissertation received the Syracuse University All-University Doctoral Prize. Prior to joining LLNL, Dr. Ray was an assistant professor at the Indian Institute of Technology (IIT), Kharagpur, India where he supervised a research group of approximately 10 scholars in the areas of statistical signal processing, wireless communications, optimization, machine learning and Bayesian non-parametrics. Prior to this he was a research scientist with the Department of Electrical and Computer Engineering at Duke University. Dr. Ray has published close to 40 research articles in various highly-rated journals and conference proceedings and is also a reviewer for leading journals in the areas of statistical signal processing, wireless communications and data science. At LLNL, Dr. Ray has been the PI/Co-I on multiple LDRDs as well as a DARPA funded research effort in the areas of machine learning for healthcare and collaborative autonomy. Host: Paul Bogdan
          Microcontroller Runs Neural Networks That Train Themselves      Cache   Translate Page      
Before machine learning algorithms can be used in factories to detect equipment malfunctions or cars to autonomously tell the difference between left and right turn arrows, they need training. That currently takes place in data centers, where neural
          What big health insurers are doing with big data      Cache   Translate Page      
Insurance companies like Kaiser, Aetna, Anthem and UnitedHealth have been making big data pays

When Vator and HP held the SplashX Invent Health - Precision Health salon at the end of September, one of the things that came up repeatedly was how big data was changing healthcare from every perspective, from the doctor to the patient to the entrepreneur. Big data is fueling everything from precision medicine to technologies like artificial intelligence and machine learning. You can't ... [Read more]

          drucker 0.4.0      Cache   Translate Page      
A Python gRPC framework for serving a machine learning module written in Python.
          Technical Architect - Data Solutions - CDW - Milwaukee, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Milwaukee, WI jobs
          Technical Architect - Data Solutions - CDW - Madison, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Madison, WI jobs
          Director, Marketing Automation for Relationship Marketing - Microsoft - Redmond, WA      Cache   Translate Page      
In addition, this role will lead the identification and leveraging of the latest marketing capabilities like machine learning and AI to propel Relationship...
From Microsoft - Wed, 31 Oct 2018 07:57:36 GMT - View all Redmond, WA jobs
          Business Analysis Manager - Business Intelligence - T-Mobile - Bellevue, WA      Cache   Translate Page      
Entrepreneurial spirit and interest in advance analytics, big data, machine learning, and AI. Do you enjoy using data to influence technology, operations and...
From T-Mobile - Wed, 10 Oct 2018 03:14:47 GMT - View all Bellevue, WA jobs
          How to Use the TimeseriesGenerator for Time Series Forecasting in Keras      Cache   Translate Page      

Time series data must be transformed into a structure of samples with input and output components before it can be used to fit a supervised learning model. This can be challenging if you have to perform this transformation manually. The Keras deep learning library provides the TimeseriesGenerator to automatically transform both univariate and multivariate time […]

The post How to Use the TimeseriesGenerator for Time Series Forecasting in Keras appeared first on Machine Learning Mastery.


          Evolution of Chat Bot      Cache   Translate Page      
Added: Nov 06, 2018
By: siyacarla1
Views: 5
What is a chatbot? A chatbot is a computer program or artificial intelligence, designed to convincingly simulate how a human would behave as a conversational partner, thereby passing the Turing test. Such programs are often designed to convincingly simulate how a human would behave as a conversational partner, thereby passing the Turing test. Chatbots are typically used in dialog systems for various practical purposes including customer service or information acquisition. Some chatterbots use sophisticated natural language processing systems, but many simpler systems scan for keywords within the input, then pull a reply with the most matching keywords, or the most similar wording pattern, from a database. Why are Chatbots Important? Chatbots are revolutionizing the way we interact. It has become an interesting junction for both human and machine learning. There had been a significant rise of adoption of chatbots in 2017 across all industries. The segment that has seen highest adoption is particularly the financial services and insurance industries.

          How machine learning is revolutionizing software development      Cache   Translate Page      
Professor Chris Bishop, director of the Microsoft Research Lab in Cambridge, UK, spells out the impact of machine learning.
          Open Source Machine Learning Tool Could Help Choose Cancer Drugs      Cache   Translate Page      
Using machine learning to analyze RNA expression tied to information about patient outcomes with specific drugs, the open source tool could help ...
          Building a rock star Data Science team      Cache   Translate Page      
It's been a year since we kicked off the IBM Data Science Elite Team. This group of expert consultants in the field of data science and machine learning ...
          AWS, Veritone: Machine Learning, AI Can Help Monetize Ads, Content      Cache   Translate Page      
Its new drone initiative is “driven by deep learning and machine learning capabilities” also, he said, noting the company has “thousands of ...
          AWS, Veritone: Machine Learning, AI Can Help Monetize Ads, Content      Cache   Translate Page      
Machine learning (ML) and artificial intelligence can be used by media and entertainment (M&E) companies to help them monetize broadcast ads and ...
          Machine learning is about to transform these industries      Cache   Translate Page      
TechRepublic's Dan Patterson asks Schneider Electric Chief Digital Officer Herve Coureil about how machine learning will transform industries.
          Artificial Intelligence Hits the Barrier of Meaning      Cache   Translate Page      
The New York Times – Machine learning algorithms don’t yet understand things the way humans do — with sometimes disastrous consequences. “…As someone who has worked in A.I. for decades, I’ve witnessed the failure of similar predictions of imminent human-level A.I., and I’m certain these latest forecasts will fall short as well. The challenge of […]
          The future of information architecture      Cache   Translate Page      
A speculative look at where Information Architecture might be going, given machine learning and voice interfaces.
          Consultant, Business Analytics & Data Science - Lincoln Financial - Boston, MA      Cache   Translate Page      
Phoenix, AZ (Arizona). Knowledge and experience on applying statistical and machine learning techniques on real business data....
From Lincoln Financial Group - Fri, 02 Nov 2018 02:54:18 GMT - View all Boston, MA jobs
          AWS Architect - Insight Enterprises, Inc. - Chicago, IL      Cache   Translate Page      
Database architecture, Big Data, Machine Learning, Business Intelligence, Advanced Analytics, Data Mining, ETL. Internal teammate application guidelines:....
From Insight - Thu, 12 Jul 2018 01:56:10 GMT - View all Chicago, IL jobs
          Amazon considers New York, Virginia amid reports of HQ split      Cache   Translate Page      

NEW YORK — After a yearlong search for a second home, Amazon is now reportedly looking to build offices in two cities instead of one, a surprise move that could still have a major impact on the communities it ultimately selects.

New York's Long Island City as well as Crystal City in northern Virginia have emerged as the front runners, according to sources familiar with the talks with Amazon.

Selecting those areas would bring more jobs to places that already have plenty. Jed Kolko, the chief economist at job site Indeed, said that choosing New York and the D.C. area would "be a much less radical move than many imagined" and another example of "rich places getting richer."

The company had originally promised to bring 50,000 new high-paying jobs to one location, which founder and CEO Jeff Bezos said would be "a full equal" to its Seattle home base. Amazon may now split those jobs equally between two locations, The Wall Street Journal reported, with each getting 25,000.

That would beg the question of whether the new locations would be headquarters at all. Kolko said a headquarters is "where the decision makers are," but it's unclear where Amazon's executives — such as Bezos — would spend much of their time. If Amazon decides to split the 50,000 workers in two places, each of those offices would be smaller than Seattle's, which has more than 40,000 employees.

Virginia officials and some state lawmakers were recently briefed by the head of the state's economic development office that Amazon was considering splitting up its second headquarters, according to a person familiar with the matter.

Officials in Virginia believe there's a strong likelihood Amazon will pick Crystal City as one of its sites, but the company has not said anything definitive, according to the person, who was not authorized to speak on the record.

"They're a real secretive company," the person said.

One of the other areas the online retail giant is considering is Long Island City, according to a source familiar with the talks. Across the East River from midtown Manhattan, Long Island City is a longtime industrial and transportation hub that has become a fast-growing neighborhood of riverfront high-rises and redeveloped warehouses, with an enduring industrial foothold and burgeoning arts and tech scenes.

Amazon has been tight-lipped about the process and declined to comment on the latest news. There's been intense competition to win over the company, with some throwing around billions of dollars in tax incentives. Amazon kicked off its hunt for a second headquarters in September 2017, initially receiving 238 proposals before narrowing the list to 20 in January.

New York Gov. Andrew Cuomo met two weeks ago with Amazon officials in his New York City offices, according to the source, who was not authorized to discuss the negotiations and spoke on condition of anonymity. Cuomo offered to travel to Amazon's Seattle hometown to continue talks, the source said.

On Tuesday, Cuomo told reporters that Amazon is looking at Long Island City, but didn't say if it was a finalist. He said winning over Amazon would give an economic boost to the entire state, and joked that he was willing to change his name to "Amazon Cuomo" to lure the company.

An estimated 135,000 or more people live in Long Island City and neighboring Sunnyside and Woodside, and the median household makes about $63,500 a year, a bit higher than the citywide median, according to New York University's Furman Center housing and urban policy think tank. About 40 percent of people over 25 in the Long Island City area have a bachelor's or higher degree, slightly above the citywide rate, the Furman Center's data shows.

The New York Times reported Monday that Amazon is finalizing deals to locate to Long Island City and the Crystal City section of Arlington, Virginia, just outside Washington, D.C. The Wall Street Journal, which first reported on the possible plan to split the headquarters between two cities, said Dallas is also still a contender. Both newspapers cited unnamed people familiar with the decision-making process.

A spokesman for the Dallas Regional Chamber declined to comment.

Long Island City and Crystal City would meet Amazon's requirements for a new locale: Both are near metropolitan areas with more than a million people, have nearby international airports, direct access to mass transit and have room for the company to expand.

Other locations that were on Amazon's list of 20 either declined to comment or said they haven't heard from the online retailer.

Jay Ash, the economic development chief in Massachusetts, said Tuesday that he's had "no recent contact" with Amazon about a headquarters in Boston, but his office is still talking with the company about other opportunities. Earlier this year, Amazon unveiled plans for an office expansion in Boston's Seaport District, promising 2,000 new technology jobs by 2021 in fields including machine learning and robotics.

Amazon has said it could spend more than $5 billion on the new headquarters over the next 17 years, about matching the size of its current home in Seattle, which has 33 buildings and 23 restaurants.

The company already employs more than 600,000 worldwide. That's expected to increase as it builds more warehouses across the country to keep up with online orders. Amazon recently announced that it would pay all its workers at least $15 an hour, but the employees at its second headquarters will be paid a lot more — an average of more than $100,000 a year.

Earlier this month, Bezos said during an on-stage interview in New York that the final decision will come down to intuition.

"You immerse yourself in that data, but then you make that decision with your heart," he said.

___

Klepper reported from Albany, New York, and Suderman reported from Richmond, Virginia. AP Technology Writer Matt O'Brien in Providence, Rhode Island, and Terry Wallace in Dallas and Jennifer Peltz in New York also contributed to this report.

Author(s): 

Articles

Blog Posts

a94a2fc7eaac459a9ad5ff177a33509a.jpg

FILE - This Sept. 6, 2012, file photo, shows the Amazon logo in Santa Monica, Calif. Online leader Amazon Inc. has refused comment on reports that it plans to split its new headquarters between two locations. The Wall Street Journal and New York Times reported late Monday, Nov. 5, 2018, that the company would locate the new facilities in Queens in New York City and in the Crystal City area of Arlington, Virginia. (AP Photo/Reed Saxon, File)
Source: 
AP

          یادگیری ماشین چیست؟ — به زبان ساده      Cache   Translate Page      
«یادگیری ماشین» (Machine Learning) یکی از کاربردهای «هوش مصنوعی» (Artificial Intelligence | AI) است که به سیستم‌ها توانایی آن را می‌دهد که به صورت خودکار بیاموزند و براساس تجربیات، بدون آنکه صراحتا برنامه‌ریزی شوند خود را بهبود ببخشند. یادگیری ماشین بر توسعه برنامه‌های کامپیوتری تمرکز دارد که می‌توانند به داده‌ها دسترسی داشته باشند و از […]
          AMD Announces Radeon Instinct MI60 & MI50 Accelerators: Powered By 7nm Vega      Cache   Translate Page      

As part of this morning’s Next Horizon event, AMD formally announced the first two accelerator cards based on the company’s previously revealed 7nm Vega GPU. Dubbed the Radeon Instinct MI60 and Radeon Instinct MI50, the two cards are aimed squarely at the enterprise accelerator market, with AMD looking to significantly improve their performance competitiveness in everything from HPC to machine learning.

Both cards are based on AMD’s 7nm GPU, which although we’ve known about at a high level for some time now, we’re only finally getting some more details on. GPU is based on a refined version of AMD’s existing Vega architecture, essentially adding compute-focused features to the chip that are necessary for the accelerator market. Interestingly, in terms of functional blocks here, 7nm Vega is actually rather close to the existing 14nm “Vega 10” GPU: both feature 64 CUs and HBM2. The difference comes down to these extra accelerator features, and the die size itself.

With respect to accelerator features, 7nm Vega and the resulting MI60 & MI50 cards differentiates itself from the previous Vega 10-powered MI25 in a few key areas. 7nm Vega brings support for half-rate double precision – up from 1/16th rate – and AMD is supporting new low precision data types as well. These INT8 and INT4 instructions are especially useful for machine learning inferencing, where high precision isn’t necessary, with AMD able to get up to 4x the perf of an FP16/INT16 data type when using the smallest INT4 data type. However it’s not clear from AMD’s presentation how flexible these new data types are – and with what instructions they can be used – which will be important for understanding the full capabilities of the new GPU. All told, AMD is claiming a peak throughput of 7.4 TFLOPS FP64, 14.7 TFLOPS FP32, and 118 TOPS for INT4.

7nm Vega also buffs up AMD’s memory capabilities. The GPU adds another pair of HBM2 memory controllers, giving it 4 in total. Combined with a modest increase in memory clockspeeds to 2Gbps, and AMD now has a full 1TB/sec of memory bandwidth in the GPU’s fastest configuration. This is even more than NVIDIA’s flagship GV100 GPU, giving AMD the edge in bandwidth. Meanwhile as this is an enterprise-focused GPU, it offers end-to-end ECC, marking the first AMD GPU to offer complete ECC support in several years.

The enterprise flourishes also apply to 7nm Vega’s I/O options. On the PCIe front, AMD has revealed that the GPU supports the recently finalized PCIe 4 standard, which doubles the amount of memory bandwidth per x16 slot to 31.5GB/sec. However AMD isn’t stopping there. The new GPU also includes a pair of off-chip Infinity Fabric links, allowing for the Radeon Instinct cards to be directly connected to each other via the coherent links. I’m still waiting for a confirmed breakdown on the numbers, but it looks like each link supports 50GB/sec down and 50GB/sec up in bandwidth.

Notably, since there are only 2 links per GPU, AMD’s topology options will be limited to variations on rings. So GPUs in 4-way configurations won’t all be able to directly address each other. Meanwhile AMD is still sticking with PCIe cards as their base form factor here – no custom mezzanine-style cards like NVIDIA – so the cards are connected via a bridge on the top. Meanwhile backhaul to the CPU (AMD suggests an Epyc, of course) is handled over PCIe 4.

Finally, looking at the GPU itself, it’s interesting to note just how small it is. Because AMD didn’t significantly bulk up the GPU on CUs, thanks to the 7nm process the new GPU is actually a good bit smaller than the original 484mm2 Vega 10 GPU. The new GPU comes in at 331mm2, packing in 13.2B transistors. Though it should be noted that AMD’s performance estimates are realistically conservative here; while 7nm does bring power consumption down, AMD is still only touting >1.25x performance of MI25 at the same power consumption. The true power in the new cards lies in their new features, rather than standard FP16/FP32 calculations that the existing MI25 card was already geared for.

Wrapping things up, Radeon Instinct MI60 will be shipping in Q4 of this year. AMD has not announced a price, but as a cutting-edge 7nm GPU, don’t expect it to be cheap. MI60 will then be followed by MI50 in Q1 of next year, giving AMD’s customers a second, cheaper option to access 7nm Vega.

Gallery: Rome Presentation Slide Deck


          Machine Learning Engineer - Workbridge Associates - Toronto, ON      Cache   Translate Page      
One of my best clients is hiring for Machine Learning Engineers and are willing to pay the big bucks for the right talent.... $120 - $155 a day
From Workbridge Associates - Mon, 05 Nov 2018 14:26:59 GMT - View all Toronto, ON jobs
          LXer: OPNids Integrates Machine Learning Into Open Source Suricata IDS      Cache   Translate Page      
Published at LXer: New open source project gets underway integrating the Suricata Intrusion Detection System (IDS) with the DragonFly Machine Learning Engine, which uses a streaming data analytics...
          SNR. PYTHON DEVELOPER- DEVELOP YOUR MACHINE LEARNING AND DATA ANALYTICS SKILLS      Cache   Translate Page      
Acuity Consultants - Paarl, Western Cape - This is an excellent opportunity for a SNR. PYTHON DEVELOPER to develop their machine learning and data analytics skills. Based in the... has pioneered the InsureTech space in South Africa, by capitalizing on data science and machine learning technology to create the country's first award...
          PYTHON DEVELOPER- DEVELOP YOUR MACHINE LEARNING AND DATA ANALYTICS SKILLS      Cache   Translate Page      
Acuity Consultants - Paarl, Western Cape - This is an excellent opportunity for a Python developer to develop their machine learning and data analytics skills. Based in the NORTHERN... has pioneered the InsureTech space in South Africa, by capitalizing on data science and machine learning technology to create the country's first award...
          Sr/Principal Consultant, Red Team - Cylance, Inc. - Texas      Cache   Translate Page      
Internal / External / Wireless - Penetration Testing (2+ years REQUIRED). By successfully applying artificial intelligence and machine learning to crack the DNA...
From Cylance, Inc. - Wed, 05 Sep 2018 19:27:50 GMT - View all Texas jobs
          PKI Engineer - Cylance, Inc. - Texas      Cache   Translate Page      
Data exchanges with internal and external security intelligence platforms. By successfully applying artificial intelligence and machine learning to crack the...
From Cylance, Inc. - Wed, 05 Sep 2018 19:27:49 GMT - View all Texas jobs
          Red Team Consultant - Cylance, Inc. - North Carolina      Cache   Translate Page      
By successfully applying artificial intelligence and machine learning to crack the DNA of malware, Cylance has redefined the endpoint protection market,...
From Cylance, Inc. - Wed, 05 Sep 2018 01:27:47 GMT - View all North Carolina jobs
          Technical Account Manager (East Coast, South East, Midwest - REMOTE) - Cylance, Inc. - Florida      Cache   Translate Page      
By successfully applying artificial intelligence and machine learning to crack the DNA of malware, Cylance has redefined the endpoint protection market,...
From Cylance, Inc. - Fri, 27 Jul 2018 19:28:10 GMT - View all Florida jobs
          Senior Compliance Analyst - Cylance, Inc. - Irvine, CA      Cache   Translate Page      
By successfully applying artificial intelligence and machine learning to crack the DNA of malware, Cylance has redefined the endpoint protection market,...
From Cylance, Inc. - Thu, 13 Sep 2018 19:27:37 GMT - View all Irvine, CA jobs
          Senior Compliance & Privacy Analyst - Cylance, Inc. - Irvine, CA      Cache   Translate Page      
By successfully applying artificial intelligence and machine learning to crack the DNA of malware, Cylance has redefined the endpoint protection market,...
From Cylance, Inc. - Sat, 08 Sep 2018 01:27:53 GMT - View all Irvine, CA jobs
          Financial Reporting Director - Cylance, Inc. - Irvine, CA      Cache   Translate Page      
By successfully applying artificial intelligence and machine learning to crack the DNA of malware, Cylance has redefined the endpoint protection market,...
From Cylance, Inc. - Mon, 13 Aug 2018 07:27:33 GMT - View all Irvine, CA jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Seattle, WA      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 07 Sep 2018 02:02:14 GMT - View all Seattle, WA jobs
          Sr. Associate, ML Pipelines for AI Consultant - KPMG - McLean, VA      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 26 Oct 2018 15:22:01 GMT - View all McLean, VA jobs
          Sr. Associate, AI in Management Analytics Consultant - KPMG - McLean, VA      Cache   Translate Page      
Ability to apply statistical, machine learnings, and artificial intelligence techniques to achieve concrete business goals and work with the business to...
From KPMG LLP - Sat, 29 Sep 2018 15:21:53 GMT - View all McLean, VA jobs
          Data Scientist - Deloitte - Springfield, VA      Cache   Translate Page      
Demonstrated knowledge of machine learning techniques and algorithms. We believe that business has the power to inspire and transform....
From Deloitte - Fri, 10 Aug 2018 06:29:44 GMT - View all Springfield, VA jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Dallas, TX      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 07 Sep 2018 02:02:14 GMT - View all Dallas, TX jobs
          Associate, Machine Learning AI Consultant, Financial Services - KPMG - Dallas, TX      Cache   Translate Page      
Broad, versatile knowledge of analytics and data science landscape, combined with strong business consulting acumen, enabling the identification, design and...
From KPMG LLP - Fri, 07 Sep 2018 02:02:14 GMT - View all Dallas, TX jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Philadelphia, PA      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 14 Sep 2018 08:38:34 GMT - View all Philadelphia, PA jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - New York, NY      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 14 Sep 2018 08:38:34 GMT - View all New York, NY jobs
          Associate, Machine Learning AI Consultant, Financial Services - KPMG - New York, NY      Cache   Translate Page      
Broad, versatile knowledge of analytics and data science landscape, combined with strong business consulting acumen, enabling the identification, design and...
From KPMG LLP - Fri, 14 Sep 2018 08:38:34 GMT - View all New York, NY jobs
          C/C++ Software Engineer/Machine Learning - Mobica - Warszawa, mazowieckie      Cache   Translate Page      
I hereby agree for processing of personal data by Mobica Limited with headquarter in Crown House, Manchester Road, Wilmslow, UK, SK9 1BH whose representative is...
Od Mobica - Fri, 19 Oct 2018 09:05:38 GMT - Pokaż wszystkie Warszawa, mazowieckie oferty pracy
          Data Scientist: Medical VoC and Text Analytics Manager - GlaxoSmithKline - Research Triangle Park, NC      Cache   Translate Page      
Strong business acumen; 2+ years of unstructured data analysis/text analytics/natural language processing and/or machine learning application for critical...
From GlaxoSmithKline - Fri, 19 Oct 2018 23:19:12 GMT - View all Research Triangle Park, NC jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Charlotte, NC      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Wed, 17 Oct 2018 08:49:16 GMT - View all Charlotte, NC jobs
          Senior C/C++ Software Engineer/Machine Learning - Mobica - Warszawa, mazowieckie      Cache   Translate Page      
I hereby agree for processing of personal data by Mobica Limited with headquarter in Crown House, Manchester Road, Wilmslow, UK, SK9 1BH whose representative is...
Od Mobica - Fri, 14 Sep 2018 21:05:32 GMT - Pokaż wszystkie Warszawa, mazowieckie oferty pracy
          Protecting What Matters: Defining Data Guardrails and Behavioral Analytics      Cache   Translate Page      

Posted under: General

This is the second post in our series on Protecting What Matters: Introducing Data Guardrails and Behavioral Analytics. Our first post, Introducing Data Guardrails and Behavioral Analytics: Understand the Mission, introduced the concepts and outlined the major categories of insider risk. This post defines the concepts.

Data security has long been the most challenging domain of information security, despite being the centerpiece of our entire practice. We only call it “data security” because “information security” was already taken. Data security must not impede use of the data itself. By contrast it’s easy to protect archival data (encrypt it and lock the keys up in a safe). But protecting unstructured data in active use by our organizations? Not so easy. That’s why we started this research by focusing on insider risks, including external attackers leveraging insider access. Recognizing someone performing an authorized action, but with malicious intent, is a nuance lost on most security tools.

How Data Guardrails and Data Behavioral Analytics are Different

Both data guardrails and data behavioral analytics strive to improve data security by combining content knowledge (classification) with context and usage. Data guardrails leverage this knowledge in deterministic models and processes to minimize the friction of security while still improving defenses. For example, if a user attempts to make a file in a sensitive repository public, a guardrail could require them to record a justification and then send a notification to Security to approve the request. Guardrails are rule sets that keep users “within the lines” of authorized activity, based on what they are doing.

Data behavioral analytics extends the analysis to include current and historical activity, and uses tools such as artificial intelligence/machine learning and social graphs to identify unusual patterns which bypass other data security controls. Analytics reduces these gaps by looking not only at content and simple context (as DLP might), but also adding in history of how that data, and data like it, has been used within the current context. A simple example is a user accessing an unusual volume of data in a short period, which could indicate malicious intent or a compromised account. A more complicated situation would identify sensitive intellectual property on an accounting team device, even though they do not need to collaborate with the engineering team. This higher order decision making requires an understanding of data usage and connections within your environment.

Central to these concepts is the reality of distributed data actively used widely by many employees. Security can’t effectively lock everything down with strict rules covering every use case without fundamentally breaking business processes. But with integrated views of data and its intersection with users, we can build data guardrails and informed data behavioral analytical models, to identify and reduce misuse without negatively impacting legitimate activity. Data guardrails enforce predictable rules aligned with authorized business processes, while data behavioral analytics look for edge cases and less predictable anomalies.

How Data Guardrails and Data Behavioral Analytics Work

The easiest way to understand the difference between data guardrails and data behavioral analytics is that guardrails rely on pre-built deterministic rules (which can be as simple as “if this then that”), while analytics rely on AI, machine learning, and other heuristic technologies which look at patterns and deviations.

To be effective both rely on the following foundational capabilities:

  • A centralized view of data. Both approaches assume a broad understanding of data and usage – without a central view you can’t build the rules or models.
  • Access to data context. Context includes multiple characteristics including location, size, data type (if available), tags, who has access, who created the data, and all available metadata.
  • Access to user context, including privileges (entitlements), groups, roles, business unit, etc.
  • The ability to monitor activity and enforce rules. Guardrails, by nature, are preventative controls which require enforcement capabilities. Data behavioral analytics can be used only for detection, but are far more effective at preventing data loss if they can block actions.

The two technologies then work differently while reinforcing each other:

  • Data guardrails are sets of rules which look for specific deviations from policy, then take action to restore compliance. To expand our earlier example:
    • A user shares a file located in cloud storage publicly. Let’s assume the user has the proper privileges to make files public. The file is in a cloud service so we also assume centralized monitoring/visibility, as well as the capability to enforce rules on that file.
    • The file is located in an engineering team’s repository (directory) for new plans and projects. Even without tagging, this location alone indicates a potentially sensitive file.
    • The system sees the request to make the file public, but because of the context (location or tag), it prompts the user to enter a justification to allow the action, which gets logged for the security team to review. Alternatively, the guardrail could require approval from a manager before allowing the action.

Guardrails are not blockers because the user can still share the file. Prompting for user justification both prevents mistakes and loops in security review for accountability, allowing the business to move fast while minimizing risk. You could also look for large file movements based on pre-determined thresholds. A guardrail would only kick in if the policy thresholds are violated, and then use enforcement actions aligned with business processes (such as approvals and notifications) rather than simply blocking activity and calling in the security goons.

  • Data behavioral analytics use historical information and activity (typically with training sets of known-good and known-bad activity), which produce artificial intelligence models to identify anomalies. We don’t want to be too narrow in our description, because there are a wide variety of approaches to building models.
    • Historical activity, ongoing monitoring, and ongoing modeling are all essential – no matter the mathematical details.
    • By definition we focus on the behavior of data as the core of these models, rather than user activity; this represents a subtle but critical distinction from User Behavioral Analytics (UBA). UBA tracks activity on a per-user basis. Data behavioral analytics (the acronym DBA is already taken, so we’ll skip making up a new TLA), instead looks at activity at the source of the data. How has that data been used? By which user populations? What types of activity happen using the data? When? We don’t ignore user activity, but we track usage of data.
      • For example we could ask, “Has a file of this type ever been made public by a user in this group?” UBA would ask “Has this particular user ever made a file public?” Focusing on the data offers a chance potential to catch a broader range of data usage anomalies.
    • At risk fo stating the obvious, the better the data, the better the model. As with most security-related data science, don’t assume more data inevitably produces better models. It’s about the quality of the data. For example social graphs of communication patterns among users could be a valuable feed to detect situations like files moving between teams who do not usually collaborate. That’s worth a look, even if you wouldn’t want to block the activity outright.

Data guardrails handle known risks, and are especially effective at reducing user error and identifying account abuse resulting from tricking authorized users into unauthorized actions. Guardrails may even help reduce account takeovers, because attackers cannot misuse data if their action violate a guardrail. Data behavioral analytics then supplements guardrails for unpredictable situations and those where a bad actor tries to circumvent guardrails, including malicious misuse and account takeovers.

Now you have a better understanding of the requirements and capabilities of data guardrails and data behavioral analytics. Our next post will focus on some quick wins to justify including these capabilities in your data security strategy.

- Rich (0) Comments Subscribe to our daily email digest
          Simple Data entry - Upwork      Cache   Translate Page      
There are approx 7 worksheets that need formatting. About 300 names on each. Each sheet takes 20 minutes to format. Possible ongoing work for the right person. Files must be sent back in .csv format.

Budget: $15
Posted On: November 07, 2018 06:39 UTC
ID: 214651067
Category: Data Science & Analytics > Machine Learning
Skills: Data Entry
Country: Australia
click to apply
          Senior Site Reliability Engineer - Sift Science - Seattle, WA      Cache   Translate Page      
The Sift Science Trust PlatformTM uses real-time machine learning to accurately predict which users businesses can trust, and which ones they can't....
From Sift Science - Sun, 21 Oct 2018 06:15:49 GMT - View all Seattle, WA jobs
          Senior Software Development Engineer - Distributed Computing Services (Hex) - Amazon.com - Seattle, WA      Cache   Translate Page      
Knowledge and experience with machine learning technologies. We enable Amazon’s internal developers to improve time-to-market by allowing them to simply launch...
From Amazon.com - Thu, 26 Jul 2018 19:20:25 GMT - View all Seattle, WA jobs
          Machine Learning Engineer - Stefanini - McLean, VA      Cache   Translate Page      
AWS, Spark, Scala, Python, Airflow, EMR, Redshift, Athena, Snowflake, ECS, DevOps Automation, Integration, Docker, Build and Deployment Tools Ability to provide...
From Indeed - Tue, 16 Oct 2018 20:59:09 GMT - View all McLean, VA jobs
          Product Manager, Data - Upserve - Providence, RI      Cache   Translate Page      
Some of the technologies in our stack include React / React Native, Kinesis, DynamoDB, Machine Learning, RedShift, RDS, Ruby, JavaScript, Docker, ECS, and more....
From Upserve - Wed, 10 Oct 2018 18:03:02 GMT - View all Providence, RI jobs
          Executive Director- Machine Learning & Big Data - JP Morgan Chase - Jersey City, NJ      Cache   Translate Page      
We would be partnering very closely with individual lines of business to build these solutions to run on either the internal and public cloud....
From JPMorgan Chase - Thu, 01 Nov 2018 11:32:47 GMT - View all Jersey City, NJ jobs
          Sr Data Scientist Engineer (HCE) - Honeywell - Atlanta, GA      Cache   Translate Page      
50 Machine Learning. Develop relationships with business team members by being proactive, displaying a thorough understanding of the business processes and by...
From Honeywell - Thu, 20 Sep 2018 02:59:11 GMT - View all Atlanta, GA jobs
          Chief Strategist for Solutions and Data Architecture - HP - Palo Alto, CA      Cache   Translate Page      
Leverage Machine Learning/AI and NLP components in a solution. University degree in Computer Science or Engineering with great business understanding....
From HP - Tue, 16 Oct 2018 11:40:45 GMT - View all Palo Alto, CA jobs
          EECS presents annual awards for outstanding PhD and SM theses      Cache   Translate Page      

Anne Stuart | EECS

The faculty and leadership of the Department of Electrical Engineering and Computer Science (EECS) recently presented 13 awards for outstanding student work on recent master’s and PhD theses. Awards and recipients included:

Jin-Au Kong Award for Best PhD Theses in Electrical Engineering

  • Yu-Hsin Chen, now Research Scientist, NVIDIA Research, for “Architecture Design for Highly Flexible and Energy-Efficient Deep Neural Network Accelerators. Professors Vivienne Sze and Joel Emer, supervisors.
  • Chiraag Juvekar, now Research Scientist, Analog Garage, Analog Devices, for “Hardware and Protocols for Authentication and Secure Computation.” Professor Anantha Chandrakasan, Supervisor.  

George M. Sprowls Awards for Best PhD Theses in Computer Science

  • Arturs Backurs, now Research Assistant Professor, Toyota Technological Institute at Chicago (TTIC), for “Below P vs NP: Fine-Grained Hardness for Big Data Problems.” Professor Piotr Indyk, supervisor.
  • Gregory Bodwin, now Postdoctoral Researcher, Georgia Institute of Technology, for Sketching Distances in Graphs.” Professor Virginia Williams, supervisor.
  • Zoya Bylinskii, now Research Scientist, Adobe Research, for “Computational Perception for Multi-Modal Document Understanding.” Professor Fredo Durand and Dr. Aude Oliva, supervisors.
  • David Harwath, now Research Scientist, Spoken Languages Systems Group, Computer Science and Artificial Intelligence Laboratory (CSAIL), for “Learning Spoken Language Through Vision.” Dr. James R. Glass, supervisor.
  • Jerry Li, now VM Research Fellow, Simons Institute, University of California Berkeley, for “Principled Approaches to Robust Machine Learning Beyond.” Professor Ankur Moitra, supervisor.
  • Ludwig Schmdit, now Postdoctoral Researcher in Computer Science, University of California Berkeley, for “Algorithms Above the Noise Floor.” Professor Piotr Indyk, supervisor. "Mechanism Design: From Optimal Transport Theory to Revenue Maximization." Professor Constantinos Daskalakis, supervisor.
  • Adriana Schulz, now Assistant Professor, University of Washington, for “Computational Design for the Next Manufacturing Revolution.” Professor Wojciech Matusik, supervisor.

Ernst A. Guillemin Award for Best SM Thesis in Electrical Engineering

  • Matthew Brennan, now a PhD student in EECS at MIT, for “Reducibility and Computational Lower Bounds for Problems with Planted Sparce Structure.” Professor Guy Bresler, supervisor.
  • Syed Muhammad Imaduddin, now a PhD student in EECS at MIT, for “A Pseudo-Bayesian Model-Based Approach for Noninvasive Intracranial Pressure Estimation.” Professor Thomas Heldt, supervisor.

William A. Martin Award for Best SM Thesis in Computer Science

  • Favyen Bastani, now a PhD student in EECS at MIT, for “Robust Road Topology Extraction from Aerial Imagery.” Professors Sam Madden, Hari Balakrishnan, and Mohammad Alizadeh.
  • Wengong Jin, now a PhD student in EECS at MIT, for “Neural Graph Representation Learning with Application to Chemistry.” Professor Regina Barzilay, supervisor.

EECS Professor Martin Rinard and Professor Asu Ozdaglar, EECS department head, presented the awards during a luncheon ceremony. The PhD award winners were selected by Professor Dirk Englund (for electrical engineering) and Professor Vinod Vaikuntanathan (for computer science). The Sprowls Awards Committee, consisting of Professors Mohammad Alizadeh, Michael Carbin, and Julian Shun, assisted with selection of the PhD awards in computer science.

The SM awards were selected by Professor Elfar Adelsteinsson (for electrical engineering) and Professor Antonio Torralba (for computer science).

 

 

Date Posted: 

Tuesday, November 6, 2018 - 12:00pm

Card Title Color: 

Black

Card Description: 

Current and former EECS students were honored at a recent ceremony.

Photo: 

Card Wide Image: 


          OSS and Sharing Leftovers      Cache   Translate Page      
  • Mapillary launches an open-source Software Development Kit

    Mapillary, the street-level imagery platform that uses computer vision to improve maps for the likes of HERE, the World Bank, and automotive companies, launched an open-source Software Development Kit (SDK) to allow developers integrate the company’s image capture functionality into their own apps. The release makes it easier than ever for developers everywhere to incorporate a street-level capture component in their own apps with custom features, spearheading efforts to update maps and make them more detailed.

  • Open Source Machine Learning Tool Could Help Choose Cancer Drugs

    The selection of a first-line chemotherapy drug to treat many types of cancer is often a clear-cut decision governed by standard-of-care protocols, but what drug should be used next if the first one fails?

    That’s where Georgia Institute of Technology researchers believe their new open source decision support tool could come in. Using machine learning to analyze RNA expression tied to information about patient outcomes with specific drugs, the open source tool could help clinicians chose the chemotherapy drug most likely to attack the disease in individual patients.

    In a study using RNA analysis data from 152 patient records, the system predicted the chemotherapy drug that had provided the best outcome 80 percent of the time. The researchers believe the system’s accuracy could further improve with inclusion of additional patient records along with information such as family history and demographics.

  • Open source the secret sauce in secure, affordable voting tech

    The fastest, most cost-effective way to secure direct-record electronic voting machines in the United States, according to cybersecurity experts, is to stop using them. Switch to paper ballots and apply risk-limiting audits to ensure that vote tallies are conducted properly. And over the long term, consider switching to the cheaper—and more advanced and secure—voting technology that cybersecurity expert Ben Adida is dedicating his next career move to developing.

    Adida’s new company, which he publicly announced at the Context Conversations event here Monday evening, which The Parallax co-sponsored, is the nonprofit VotingWorks. The company, which currently is hosted by another nonprofit Adida declined to name in a conversation after the event that eventually will become its own 501(c)3, has one goal: to build a secure, affordable, open-source voting machine for use in general, public elections.

  • The Incorporation of Open Source
  • ADAF – an Open Source Digital Standards Repository for Africa – Launches in Kenya

    Digital assets provide an easy and more secure way of doing business and Africa is quickly benefiting from these new economies of trade.

    With over 1 billion people and a combined GDP of over $3.4 trillion, Africa has some of the fastest growing economies in the world.

    The African Continental Free Trade Area (AfCTA) which was signed by 44 African countries earlier this year is expected to boost intra-Africa trade tool by driving industrialisation, economic diversification and development across the African continent.

  • Infosys Launches Open Source DevOps Project

    Ravi Kumar, president and deputy COO for Infosys, said the global systems integrator has spent the last several years turning each element of its DevOps platform into a set of microservices based on containers that can now be deployed almost anywhere. That decision made it more practical for Infosys to then offer the up the entire DevOps framework it relies on to drive thousands of projects employing more than 200,000 developers as a single open source project, he said. That framework relies on an instance of the open source Jenkins continuous integration/continuous deployment (CI/CD) framework at its core.

  • Real World Data App: FDA Releases Open Source Code

    The US Food and Drug Administration (FDA) on Tuesday released computer code and a technical roadmap to allow researchers and app developers to use the agency’s newly created app that helps link real world data with electronic health data supporting clinical trials or registries.

    FDA said the app and patient data storage system can be reconfigured by organizations conducting clinical research and can be rebranded by researchers and developers who would like to customize and rebrand it.

    Among the features of the app is a secure data storage environment that supports auditing necessary for compliance with 21 CFR Part 11 and the Federal Information Security Management Act, so it can be used for trials under Investigational New Drug oversight.

  • Texas State should switch to open source textbooks
  • Big Boost For Open Access As Wellcome And Bill & Melinda Gates Foundation Back EU's 'Plan S'

    Although a more subtle change, it's an important one. It establishes unequivocally that anyone, including companies, may build on research financed by Wellcome. In particular, it explicitly allows anyone to carry out text and data mining (TDM), and to use papers and their data for training machine-learning systems. That's particularly important in the light of the EU's stupid decision to prevent companies in Europe from carrying out either TDM or training machine-learning systems on material to which they do not have legal access to unless they pay an additional licensing fee to publishers. This pretty much guarantees that the EU will become a backwater for AI compared to the US and China, where no such obstacles are placed in the way of companies.

  • Jono Bacon: Video: 10 Avoidable Career Mistakes (and How to Conquer Them)
  • 10 avoidable career mistakes (and how to conquer them)

read more


          Fastest Analytics Using Hybrid Architectures with Machine Learning      Cache   Translate Page      

Click to learn more about author Henry Bequet. Every year scientists and researchers gather at a conference called Super Computing, or SC, to exchange their views, solutions and problems in computational science. At SC17, there were no fewer than 22 presentations and keynotes involving Machine Learning and Deep Learning (DL). There were actually many more presentations about […]

The post Fastest Analytics Using Hybrid Architectures with Machine Learning appeared first on DATAVERSITY.


          LexisNexis: Removing the Hype from Big Data      Cache   Translate Page      

According to a recent press release, “LexisNexis®, a leading provider of information and analytics, has launched Nexis® Data as a Service (DaaS) to help companies power predictive analytics, machine learning and data initiatives to answer business-critical questions. Big data technologies are only as effective as the data that powers them. Organizations that tap into more […]

The post LexisNexis: Removing the Hype from Big Data appeared first on DATAVERSITY.


          Strategic Account Leader - Arundo Analytics - Houston, TX      Cache   Translate Page      
Our products help business leaders and operations professionals in heavy industries better manage complex physical systems through machine learning and data...
From Arundo Analytics - Wed, 24 Oct 2018 16:25:58 GMT - View all Houston, TX jobs
          EPO Issues First Guidelines on AI Patents      Cache   Translate Page      
The European Patent Office (EPO) has issued official guidelines on the patenting of artificial intelligence and machine learning technologies. The guidelines became valid on November 1st, 2018.
          Codeless and ML-Based Automation vs. Traditional Test Automation      Cache   Translate Page      
There’s no doubt that the test automation space is undergoing transformation. Machine Learning (ML), Deep Learning and Artificial Intelligence (AI) are being leveraged more and more as part of the test authoring and test analysis. While the space is still growing from a maturity stand-point, it is a great time for practitioners (developers and test ...
          Leveraging AI and Machine Learning to Fight Financial Crime      Cache   Translate Page      
Money laundering schemes were originally created by organized crime to “clean” money made during U.S. Prohibition in the 1930s. Law enforcement caught on and laws...
          Technical Architect - Data Solutions - CDW - Milwaukee, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Milwaukee, WI jobs
          Technical Architect - Data Solutions - CDW - Madison, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Madison, WI jobs
          Director, Marketing Automation for Relationship Marketing - Microsoft - Redmond, WA      Cache   Translate Page      
In addition, this role will lead the identification and leveraging of the latest marketing capabilities like machine learning and AI to propel Relationship...
From Microsoft - Wed, 31 Oct 2018 07:57:36 GMT - View all Redmond, WA jobs
          Business Analysis Manager - Business Intelligence - T-Mobile - Bellevue, WA      Cache   Translate Page      
Entrepreneurial spirit and interest in advance analytics, big data, machine learning, and AI. Do you enjoy using data to influence technology, operations and...
From T-Mobile - Wed, 10 Oct 2018 03:14:47 GMT - View all Bellevue, WA jobs
          URL      Cache   Translate Page      


The Expanse Is Back With a Quick, Tantalizing Look at Its Fourth Season: Oh yeah! A melhor série de FC da atualidade sobreviveu ao desinteresse do SyFy channel e mantém-se em continuidade. Está na hora de ir à descoberta do que está para lá do portal no sistema solar (spoilers: terá o seu quê de western, com monstros e artefatos alienígenas). Aqui há dias li de relance o porquê do canal ter largado o seu melhor produto: o contrato não incluía os direitos de streaming, que são agora a componente mais lucrativa do modelo financeiro televisivo.

TWENTY FIVE OF OUR ALL-TIME FAVORITE BOOKS: Uma lista menos interessante do que esperava. Alguns dos clássicos do cyberpunk, entre Sterling, Stephenson e Gibson, algum comentário mais profundo sobre tecnologia, mas o resto é mercados, mercados, mercados. Foi por isto que deixei de lera a Wired em revista, quando o deslumbre superficial pelo hype passou a ser a sua imagem de marca.

Puppy Cuteness Is Perfectly Timed to Manipulate Humans: A fofura dos cachorros como estratégia de sobrevivência, garantindo uma ligação com humanos que lhes permita sobreviver. Como dog person que sou, diria que funcionou.


The first woman Doctor Who to be commemorated with a limited-edition Barbie: Ok, esta não estava a imaginar. Claramente, a produção de Doctor Who anda à procura de outros públicos-alvo. À data em que escrevo estas linhas, a nova temporada acabou de estrear e ainda não vi o primeiro episódio com a nova Doctor, mas os zunzuns que ouvi nas redes dão esperança na série.


BruceS: Functionally Automatic?

Godmother of intelligences: Frankenstein enquanto metáfora, especialmente acutilante na era da robótica e inteligência artificial.



Celebrate NASA’s 60th birthday with these vintage photos from space: Nunca me canso de ver fotos vintage da Terra vista da Lua.

Looking back at Google+: Agora que a Google decidiu declarar defunta a rede social que ninguém usa, diria que acho que passei mais tempo a ler este artigo sobre a história da g+ (que sabe a elegia) do que realmente a usar a rede social...

Rascunhos na Voz Online – Rogério Ribeiro (Fórum Fantástico): Nesta edição, o convidado é o sensei do Fórum Fantástico.

Isto se calhar só lá vai com um empurrãozinho - Grupo LeiTugas: Nada má esta ideia do Jorge Candeias. Eu até alinho, resta saber se invento tempo para isso.

The Autocracy App: Por vezes penso que damos demasiada importância ao papel das redes sociais. Afinal, há vida para lá das apps. Depois lembro-me que vivemos numa realidade mediada por ecrãs (como anteriormente o era pelo ecrã de televisão ou página de jornal) e o quanto as redes sociais se tornaram espaços de discussão e partilha de informação. E, também, como parecem refletir o pior do ser humano, com enviesamentos, exposição pública de ódios, xenofobias, misoginias... creio que o que verdadeiramente nos entristece nas redes sociais é perceber o quão imperfeita é a humanidade. Os resmungos usam a tecnologia são mero bode expiatório para sentimentos de profunda descrença face ao ser humano.

Instagram now uses machine learning to detect bullying within photos: Boa ideia? Má ideia? Por um lado, isto cabe no tipo de filtros de conteúdos que a infame proposta de lei europeia de direitos de autor online quer implementar, embora diria que aqui os propósitos são mais nobres. Por outro lado, ajuda a detetar automaticamente casos graves de bullying. O problema aqui é que me parece que a única coisa que faz é detetar e bloquear conteúdo. Resolve o problema da imagem do Instagram enquanto rede social segura, mas não o problema mais grave do bullying.
          Hidden Costs of IoT Vulnerabilities      Cache   Translate Page      

IoT devices have become part of our work and personal lives. Unfortunately, building security into these devices was largely an afterthought.

Another day, another hack. Whether it's a baby monitor used to spy on mother and child, or anFBI warning to reset home wireless routers due to Russian intrusion, the question continues to be: What's next?

Internet of Things (IoT) devices are part of both our work and personal lives. Unfortunately, building security into these devices was largely an afterthought ― the ramifications of which we are now seeing on a near-daily basis. However, let's look beyond the headlines at the hidden costs of IoT security vulnerabilities. These fall into five categories: device security, intellectual property (IP) protection, brand protection, operational cost containment, and user experience.

Device Security

Once hacked, some devices can do a disproportionate amount of physical damage. It all depends on the degree of criticality to the nation-state, community, or individual.

The agriculture industry, for example, is as valuable to a country as any other strategic asset, such as utilities, finance, or communications. Many big farms today are automated via field sensors and autonomous vehicles. Let's imagine that someone hacks the sensors to erroneously indicate that the corn is ready to be cut, even though it's three months too early. Or that a hack signals an autonomous tractor to spread too much fertilizer, burning and causing the loss of an entire crop. This potentially catastrophic hack, as well as the corresponding financial losses or risk to the nation-state and its citizens, seem endless.

It is highly recommended that you closely examine the security of your IoT devices via the lens of worst-case scenarios. Ensuring the integrity of the data coming from your remote sensors is especially important because this data drives automated decisions with long-term implications.

IP Protection

It's astounding how many organizations will spend millions of dollars on R&D and then put that valuable intellectual property on an insecure IoT device. In this case, a hack could mean the end of your business.

Now, let's presume that you are investing heavily in building sophisticated algorithms to enable machine learning, artificial intelligence, or facial recognition. As you look to deploy these proprietary algorithms for use in an IoT device, you are ultimately left with two choices: 1) Protect the algorithm in the cloud, forcing the IoT device to run back-and-forth to run the process and adversely affecting the customer experience, or 2) install the algorithm into the OS stack on the IoT device and risk a hack that steals your algorithm ― essentially making you toss your entire R&D investment into the wastebasket.

Brand Protection

Apathy and inertia are creating a sense of "hack numbness," though the consequence of turning a blind eye depends on where you sit.

Let's say you make devices that help protect or enhance the life of children, with cameras or microphones that are always on and always watching. Consider a hack on these devices, and the misuse of the information they have access to, now being consumed by unsavory characters.

This is a brand killer. No matter how noble your IoT device and its application, if you cannot protect children, the market will make sure your future is cut short.

Consequently, security can't be ignored because you became numb to attacks. This is especially true if you're in a business that requires your IoT devices to gather sensitive information. Couple this with an emotionally invested customer base, such as users of child-monitoring devices, and a hack will mean the end of your business.

Operational Cost Containment

Satellite time is expensive. Within the broadest construct of the many new IoT devices, some will have a component that relies on satellites for data communication. It does not need to be said (but I'll say it anyway) that satellite time is a very expensive path for data backhaul.

Imagine a hack where a botnet starts a distributed denial-of-service attack on a music-streaming server, which then causes the IoT device to start rapidly and overwhelmingly pinging the music streaming service. As the IoT device is battery powered and using satellite for its backhaul, every ping now statistically shorts the life of the IoT device.

This scenario serves as a double whammy of cost containment. If you're leveraging satellites in your IoT strategy, you must examine where potential vulnerabilities are because they could affect your overall costs of operation and maintenance.

User Experience

As the saying goes, everyone has been hacked, but there are some who don't know it yet. While there may be no disruption of service at the time of a hack, what happens when there is some type of glitch?

Let's imagine that you get up one morning and ask Alexa to open the blinds, but they don't open. Now you have to check if there's Internet service into the house, and then confirm that the Wi-Fi network is broadcasting and that Alexa is enabled properly, and, finally, you have to ensure that the app for "my blinds" is connected and working. Considering how much time this could take, it would be quicker to get out of bed and just open the blinds manually.

Consequently, adding a path to ensure that the original code base is not corrupted through attestation, we can minimize the impact on the user with a highly secure device update, but the hidden cost is the impact on their time.

Conclusion

The world is catching on to the idea that IoT device security is of paramount importance. Frankly, if end users were affected in a meaningful way (say, something involving their TVs) through one significant hack, the demand for security would become "top of mind." The question is how many of these hidden costs will affect organizations while we work toward a more secure ecosystem.

In my opinion, embedding security in the IoT ecosystem can't come soon enough.

Related Content: 7 Serious IoT Vulnerabilities IoT Bot Landscape Expands, Attacks Vary by Country New Report: IoT Now Top Internet Attack Target A Cybersecurity Weak Link: linux and IoT
Hidden Costs of IoT Vulnerabilities

Black Hat Europe returns to London Dec. 3-6, 2018, with hands-on technical Trainings, cutting-edge Briefings, Arsenal open-source tool demonstrations, top-tier security solutions, and service providers in the Business Hall. Click for information on the conference and to register.

Carl Nerup's experience is a powerful mix of proven marketing and sales leadership an
          Protecting What Matters: Defining Data Guardrails and Behavioral Analytics      Cache   Translate Page      
Posted under: General

Title: Protecting What Matters: Defining Data Guardrails and Behavioral Analytics

This is the second post in our series on Protecting What Matters: Introducing Data Guardrails and Behavioral Analytics. Our first post, Introducing Data Guardrails and Behavioral Analytics: Understand the Mission we introduced the concepts and outlined the major categories of insider risk. In this post we define the concepts.

Data security has long been the most challenging domain of information security despite it being the charter of our entire practice. We only call it “data security” because “information security” was already taken. Data security cannot impede the use of the data itself. By contrast, it’s easy to protect archival data (encrypt it and lock up the keys in a safe). But protecting unstructured data in active use by our organizations? Not so easy. That’s why we started this research by focusing on insider risks, including external attackers leveraging insider access. Determining someone doing an authorized action, but with malicious intent is a nuance lost on most security tools.

How Data Guardrails and Data Behavioral Analytics are Different

Both data guardrails and data behavioral analytics strive to improve data security by combining content knowledge (classification) with context and usage. Data guardrails leverage this knowledge in deterministic models and processes to minimize the friction of security without still improving defenses. For example, if a user attempts to make a file in a sensitive repository public, a guardrail could require them to record a justification and then send a notification to security to approve the request. Guardrails are rule sets that keep users “within the lines” of authorized activity, based on what they are doing.

Data behavioral analytics extends the analysis to include current and historical activity and uses tools like artificial intelligence/machine learning and social graphs to identify unusual patterns that bypass other data security controls. They reduce these gaps by not only looking at content and simple context (as DLP might), but by adding in the history of how that data, and data like it, has been used within the current context. A simple example is a user accessing an unusual volume of data in a short period, which could indicate malicious intent or a compromised account. A more complicated situation would identify sensitive intellectual property on an accounting team device, even though they do not need to collaborate with the engineering team. This higher order decision making requires an understanding of data usage and connections within your environment.

Central to these concepts is the reality of distributed data actively used widely by many employees. Security can’t effectively lock everything down with strict rules to cover every use cases without fundamentally breaking business process. But with integrated views of data and its intersection with users, we can build data guardrails and informed data behavioral analytical models to identify and reduce misuse without negatively impacting legitimate activities. Data guardrails enforce predictable rules aligned with authorized business processes, while data behavioral analytics look for edge cases and less predictable anomalies.

How Data Guardrails and Data Behavioral Analytics Work

The easiest way to understand the difference between data guardrails and data behavioral analytics is that guardrails rely on pre-built deterministic rules (which can be as simple as “if this then that”), while analytics relies on AI, machine learning, and other heuristic-based technologies that look at patterns and deviations.

To be effective, both rely on the following foundational capabilities:

* A centralized view of the data. Both approaches assume a broad understanding of data and usage; without a central view, you can’t build the rules or models.

* Access to data context. Context includes multiple characteristics, including location, size, data type (if available), tags, who has access, who created the data, and all available metadata.

* Access to user context, including privileges (entitlements), groups, roles, business unit, etc.

* The ability to monitor activity and enforce rules. Guardrails, by nature, are preventative controls and require enforcement capabilities. Data behavioral analytics can be technically only for detection but are far more effective in preventing loss if they can block actions.

The two technologies then work differently while reinforcing each other:

Data guardrails are sets of rules that look for specific deviations from policy, then take action to restore compliance with the policy. To expand our earlier example: A user shares a file located in cloud storage publicly. Let’s assume the user has the proper privileges to make files public. Since the file is in a cloud service, we also assume centralized monitoring/visibility, as well as the capability to enforce rules on that file. The file is located in an engineering team’s repository (directory) for new plans and projects. Even without tagging, this location alone indicates a potentially sensitive file. The system sees the request to make the file public, but because of the context (location or tag), it prompts the user to enter a justification to allow the action, which gets logged for the security team to review. Alternatively, the guardrail could require approval from a manager before allowing the file action.

Guardrails are not blockers because the user can still share the file. Prompting for user justification both prevents mistakes and loops in security review for accountability, allowing the business to move fast while still minimizing risk. You could also look for large file movements based on pre-determined thresholds. A guardrail would only kick in if the policy thresholds are violated, and then use enforcement actions aligned with the business process (like approvals and notifications) rather than just blocking activity and calling in the security goons.

Data behavioral analytics use historical information and activity (typically with training sets of known-good and known-bad activity) which build artificial intelligence models identifying anomalies. We don’t want to be too narrow here in our description since there are a wide variety of approaches to building models. Historical activity, ongoing monitoring, and ongoing modeling are essential no matter the mathematical details. By definition we focus on the behavior of the data as the core of the models, not user activity, representing a subtle, but critical distinction from User behavioral analytics (UBA) . UBA tracks activity on a per-user basis. Data behavioral analytics (since the acronym DBA is already taken we’ll skip making up a new TLA), instead looks at activity at the source of the data. How has that data been used? Which user populations? What types of activity happen using the data? When? Not that we ignore user activity, but we are tracking usage of the data . For example, we are answering the question “has a file of this type ever been made public by a user in this group?” UBA would ask “has this particular user ever made a file public?” We believe focusing on the data has the potential to catch a broader range of data usage anomalies. To state the obvious, the better the data, the better the model. As with most security-related data science, don’t assume more data results in better models. It’s about the quality of the data. For example, social graphs of communications patterns among users could be a valuable feed to detect situations like files moving between teams not usually collaborating. That’s worth a look, even if you don’t want to block the activity outright.

Data guardrails handle known risks and are especially effective in reducing user error and identifying account abuse resulting from tricking authorized users into unauthorized actions. Guardrails may even help reduce account takeovers since the attackers wouldn’t be able to misuse the data if the action violated a guardrail. Data behavioral analytics then supplements the guardrail for those unpredictable situations or where the bad actor will try to circumvent the guardrails, including malicious misuse and account takeovers.

Now you have a better understanding of the requirements and capabilities of data guardrails and data behavioral analytics. In our next post, we will focus on some quick wins to justify including these capabilities in your data security strategy.

Rich

(0) Comments

Subscribe to our daily email digest
          Residence In: NVIDIA Launches Year-Long Research Residency Program      Cache   Translate Page      

If you're a researcher looking to deepen your exposure to AI, NVIDIA invites you to apply to its new AI Research Residency program. During the one-year, paid program, residents will be paired with an NVIDIA research scientist on a joint project and have the opportunity to publish and present their findings at prominent research conferences such as CVPR, ICLR and ICML. The residency program is meant to encourage scholars with diverse academic backgrounds to pursue machine learning re...

Read the full story at https://www.webwire.com/ViewPressRel.asp?aId=230885


          OPNids Integrates Machine Learning Into Open-Source Suricata IDS      Cache   Translate Page      
EXCLUSIVE: A new open-source project integrates the Suricata intrusion detection system with the DragonFly Machine Learning Engine, which uses a streaming data analytics model to help make decisions.
          Microsoft Azure Machine Learning and Project Brainwave – Intel Chip Chat – Episode 610      Cache   Translate Page      
In this Intel Chip Chat audio podcast with Allyson Klein: In this interview from Microsoft Ignite, Dr. Ted Way, Senior Program Manager for Microsoft, stops by to talk about Microsoft Azure Machine Learning, an end-to-end, enterprise grade data science platform. Microsoft takes a holistic approach to machine learning and artificial intelligence, by developing and deploying [...]
          Data Scientist      Cache   Translate Page      
TX-Irving, Hands on in applying data mining techniques, doing statistical analysis, machine learning algorithms and building high quality prediction systems integrated with products and processes. Should have experience in some of the follwoing: “automate scoring using machine learning techniques”, “build recommendation systems”, “improve and extend the features used by our existing classifier”, “develop int
          Google and Harvard develop AI to find restaurants that could make you sick      Cache   Translate Page      
A study led by researchers at the Mountain View company and Harvard’s T.H. Chan School of Public Health describes a machine learning model that leverages search and location data to identify “potentially unsafe” restaurants.
          Director, Marketing Automation for Relationship Marketing - Microsoft - Redmond, WA      Cache   Translate Page      
In addition, this role will lead the identification and leveraging of the latest marketing capabilities like machine learning and AI to propel Relationship...
From Microsoft - Wed, 31 Oct 2018 07:57:36 GMT - View all Redmond, WA jobs
          Create a functional Rasa UI chatbot demo from a call script example -- 2      Cache   Translate Page      
Create a functional Rasa UI chatbot demo from a phone call script To win this project you should have intermediate or expert knowledge of Rasa NLU and Rasa Core, using a Linux operating system on a server... (Budget: $250 - $750 USD, Jobs: Machine Learning, Natural Language, Software Architecture)
          The EU’s border control ‘lie detector’ AI is hogwash      Cache   Translate Page      

Calling the EU’s new border control AI a “lie detector” is like calling Brexit a minor disagreement among friends. The low-down is that the EU is testing a pilot program for international airports featuring a machine learning-based “lie detector.” CNN broke the story last week in its article “Passengers to face AI lie detector tests at EU airports.” According to the report (and the project’s website) the EU is testing a pilot program involving AI that uses an avatar to ask people questions. Supposedly this AI-powered construct knows if a person is being truthful when they answer, and it flags…

This story continues at The Next Web
          Analyzing expense receipts with Azure Cognitive Services and Microsoft Flow      Cache   Translate Page      
Recently, Business Applications MVP Steve Endow and I delivered a session at the User Group Summits in Phoenix, and in particular, to the GPUG Summit titled, "Microsoft Dynamics GP and Azure Services". In this course we detailed a number of Azure Services (among the hundreds) that could potentially be used with Microsoft Dynamics GP.

Being that I have also been working my way through completing a Microsoft sanctioned PowerApps, Flow, and CDS self-paced, online training class offered by edX (click here for more info on the course) and presented by Business Applications MVP Shane Young, I asked myself, what could I do with Microsoft Flow and Azure Services that could benefit Microsoft Dynamics GP users?

Playing with some Azure Services, I came across Azure Cognitive Services which offers the capability of reading images and analyzing them for text via its Computer Vision service. As it turns out, this service offers an optical character recognition (OCR) feature, which is capable of returning the full text or a JSON formatted document with the text. The idea here would be to use Microsoft Flow to read a newly created receipt in a OneDrive folder and transfer the file to Cognitive Services' Computer Vision for analysis, then get back the parsed text from the OCR service. 

Let's see how it's done!

Provision the Computer Vision service

The first thing is to ensure Computer Vision has been enabled as a service on your Azure tenant. For this visit the Azure Portal, click on Create a Resource, then select Computer Vision from the AI + Machine Learning category within the Azure Marketplace. 

Computer Vision

Fill in some basic information like the resource name, location, pricing tier (there's a F0 free service!), and a resource group. Click the Create button when done. This will provision your Computer Vision resource.
 
Copy the service endpoint address and access keys

Once the service is provisioned, click on All Resources, select the Computer Vision resource you just created, then click on Overview.

Grab the service endpoint address and the access keys. You can obtain the access keys by clicking on Show access keys.. (two access keys are provided). 

Computer Vision service endpoint info

This is, by far, one of the easiest services to provision and requires no extra configuration, beyond establishing some access control to limit who can use the service, but that's not a topic for this discussion.

Setup a new Microsoft Flow flow

Over in Microsoft Flow, I started by setting up a blank flow and selected OneDrive's "When a file is created trigger" as this would setup the simple trigger point for when an expense receipt file is added to a Receipts folder I had previously created. You will then be prompted to setup the connection information for OneDrive.

Blank flow with "When a file is created" trigger

NOTE: I selected my personal OneDrive for this, but this can also be done with a folder on your OneDrive for Business environment. In this case, you will want to authenticate with your Office 365 credentials.
Receipts folder


Submit file to Computer Vision service

As it also turns out, Microsoft Flow has a connector to the Azure Computer Vision API, which exposes two actions: OCR to JSON and OCR to Text. Add a New Step and type Computer Vision in the search bar. Select Computer Vision API, then choose OCR to Text action.

Computer Vision API connector - OCR to Text action

Once again, you will be prompted for the connection information to your Computer Vision service on Azure. Enter the user account, the access key and service endpoint as gathered in step two, above.

Computer Vision API - credentials entry
Once credentials are entered, you can decide what to submit to Computer Vision. In this case, we what to send the File Content, which we can select from Dynamic content fields.

File Content from Dynamics content fields

Configure Email step with Results

Upon completion, we want to send the resulting OCR from the analyzed image via email, so we will add another step to the flow. This time, we will a connector to Office 365 Outlook and will choose the Send an Email action for our next step.
Office 365 Outlook connector - Send Email action

We can then setup the properties for the Send an Email step. I have chosen to send the email to myself, and compose a subject line using the File name from the OneDrive directory. As body, I am including the Detected Text, which can be selected from the OCR to Text category under Dynamic content. I've included both the original file and content as part of the attachments.



Finally, I have given this flow a name and saved it.

Testing the Flow

I have dropped two receipt files within my OneDrive Receipts folder. These two receipts present various degrees of quality and text that can be recognized by the service. I was particularly interested in the second receipt (right) as this one was very folded and cracked so I was curious to see how it would be analyzed.

Receipts
For the second receipt, the OCR service returned the JSON payload and a status 200, indicating it was successful in processing and delivering a response.

JSON payload for second receipt

The actual email I received look like this and contained the following text:

Receipt analysis

Now, remember that my goal isn't to judge the accuracy of the OCR result delivered by Computer Vision, but rather to show how easy it is to build these kinds of solutions with Microsoft Flow and existing Azure services. Something like this would take an inordinate amount of time to build using traditional development tools and services.

Conceivably, I could create a simple PowerApps application that uses the Camera control to take the picture of the receipt and save it to my OneDrive folder. At this point, the receipt would be picked up by the Flow and analyzed by Computer Vision as we have established here. Why would this be important? Perhaps if you want to parse the JSON payload and rather submit to Microsoft Dynamics GP or Dynamics 365 as an account payables voucher, this would be useful.

Until next post,

MG.-
Mariano Gomez, MVP


          Vice President, Data Science - Machine Learning - Wunderman - Dallas, TX      Cache   Translate Page      
Goldman Sachs, Microsoft, Citibank, Coca-Cola, Ford, Pfizer, Adidas, United Airlines and leading regional brands are among our clients....
From Wunderman - Sat, 25 Aug 2018 05:00:40 GMT - View all Dallas, TX jobs
          Program Manager, Services - CloudMoyo - Bellevue, WA      Cache   Translate Page      
CloudMoyo’s proven track record includes developing impactful solutions using big data, machine learning, predictive analytics and visual story-telling for...
From CloudMoyo - Fri, 05 Oct 2018 23:54:14 GMT - View all Bellevue, WA jobs
          Lead/Sr. Functional Consultant, Services - CloudMoyo - Bellevue, WA      Cache   Translate Page      
CloudMoyo’s proven track record includes developing impactful solutions using big data, machine learning, predictive analytics and visual story-telling for...
From CloudMoyo - Fri, 05 Oct 2018 23:54:13 GMT - View all Bellevue, WA jobs
          Do annotation for my dataset      Cache   Translate Page      
I would like to hire a freelancer to do annotations to my image dataset (total number of images is less than 100) using VIA 1.0 (Budget: ₹600 - ₹1500 INR, Jobs: Algorithm, Artificial Intelligence, C Programming, Machine Learning, Python)
          Three ways to avoid bias in machine learning      Cache   Translate Page      

Vince Lynch Contributor Vince Lynch is CEO of IV.AI, an artificial intelligence company that teaches machines how to understand human language so companies can better engage, understand and serve their customers. At this moment in history it’s impossible not to see the problems that arise from human bias. Now magnify that by compute and you start […]

The post Three ways to avoid bias in machine learning appeared first on RocketNews | Top News Stories From Around the Globe.


          Do annotation for my dataset      Cache   Translate Page      
I would like to hire a freelancer to do annotations to my image dataset (total number of images is less than 100) using VIA 1.0 (Budget: ₹600 - ₹1500 INR, Jobs: Algorithm, Artificial Intelligence, C Programming, Machine Learning, Python)
          What is the Difference between Machine Learning and Deep Learning?      Cache   Translate Page      

There is often a lot of confusion around the differences between machine learning and deep learning.  Both are classed as techniques to enable artificial intellgence or AI.  But what is AI? AI is the ability to create a program or computer system that can fool a human into thinking it is another human.  There is […]

The post What is the Difference between Machine Learning and Deep Learning? appeared first on ALC Training.


          Three ways to avoid bias in machine learning      Cache   Translate Page      
At this moment in history it’s impossible not to see the problems that arise from human bias. Now magnify that by compute and you start to get a sense for just how dangerous human bias via machine learning can be.
          Grid4C lands $5 mln      Cache   Translate Page      
Austin, Texas-based Grid4C, a provider of AI and machine learning solutions for the energy industry, has raised $5 million in funding. ICV led the round.
          2018 Pinnacle Awards: Intel’s Melvin Greer Takes Home Artificial Intelligence Executive of the Year      Cache   Translate Page      
Melvin Greer, the chief data scientist at Intel Corp. who’s, helping chart the future of artificial intelligence in his role at the company, was awarded with an inaugural WashingtonExec Pinnacle Award for the AI Executive of the Year. Greer has drawn accolades for his work building Intel’s data science platform using AI and machine learning. [...]
          “Statistical and Machine Learning forecasting methods: Concerns and ways forward”      Cache   Translate Page      

Roy Mendelssohn points us to this paper by Spyros Makridakis, Evangelos Spiliotis, and Vassilios Assimakopoulos, which begins: Machine Learning (ML) methods have been proposed in the academic literature as alternatives to statistical ones for time series forecasting. Yet, scant evidence is available about their relative performance in terms of accuracy and computational requirements. The purpose […]

The post “Statistical and Machine Learning forecasting methods: Concerns and ways forward” appeared first on Statistical Modeling, Causal Inference, and Social Science.


          Chinese Company Tencent Enters The Self-Driving Car Race      Cache   Translate Page      

Visitors and exhibitors network at the Tencent booth during the 2016 Sportel Asia Conference in Singapore.

Chinese Technology company Tencent Holdings Ltd is reportedly entering into the self-driving car industry.

According to Reuters, Tencent is recruiting self-driving car engineers in Palo Alto, California, essentially joining companies that compete for talents in the heart of the Silicon Valley, which has become a hub for testing and research of autonomous vehicles.

Five dozen companies have permit to test these cars on the roads of California, but as of last month, state records did not show Tencent as having an autonomous vehicle testing permit.

Based on a job posting on Linkedin, however, Tencent is throwing its hat into the self-driving vehicle industry as it hunts for talents to work on the technology.

The company has at least nine postings for engineering positions in areas that include sensor fusion, vehicle intelligence, motion planning, and machine learning.

“We are building a research team for our Auto-drive Team based in Palo Alto, CA,” the company said in a job advertisement, which was posted a month ago.

Click here to continue and read more...


          Vier wichtige Punkte zum Thema maschinelles Lernen von der SAP TechEd 2018      Cache   Translate Page      

In seinem Blog erläutert Marcus Noga, Leiter des Bereichs Machine Learning bei SAP, welche Neuerungen auf der SAP TechEd vorgestellt wurden und wie Unternehmen maschinelles...

The post Vier wichtige Punkte zum Thema maschinelles Lernen von der SAP TechEd 2018 appeared first on SAP News Center.


          Python Developer/Data Scientist - RiverPoint - Houston, TX      Cache   Translate Page      
We are looking for individuals to fill the role of Data Scientist on our model development team. This team builds the machine learning algorithms that...
From RiverPoint - Sat, 03 Nov 2018 06:30:32 GMT - View all Houston, TX jobs
          Six West Coast Tech Leaders Share Top Developments of 2017      Cache   Translate Page      
Xconomy asked technology and innovation leaders around our network to reflect on the most important developments in their industries during 2017, and the answers were appropriately wide-ranging. Responses from individuals in Seattle and San Diego touch on the rapid advance of machine learning, tech’s full-scale invasion of digital health, dramatic growth in blockchain and cryptocurrency, […]
          Technical Architect - Data Solutions - CDW - Milwaukee, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Milwaukee, WI jobs
          Technical Architect - Data Solutions - CDW - Madison, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Madison, WI jobs
          Director, Marketing Automation for Relationship Marketing - Microsoft - Redmond, WA      Cache   Translate Page      
In addition, this role will lead the identification and leveraging of the latest marketing capabilities like machine learning and AI to propel Relationship...
From Microsoft - Wed, 31 Oct 2018 07:57:36 GMT - View all Redmond, WA jobs
          Business Analysis Manager - Business Intelligence - T-Mobile - Bellevue, WA      Cache   Translate Page      
Entrepreneurial spirit and interest in advance analytics, big data, machine learning, and AI. Do you enjoy using data to influence technology, operations and...
From T-Mobile - Wed, 10 Oct 2018 03:14:47 GMT - View all Bellevue, WA jobs
          Kienbaum discute sobre o tema “trabalho 4.0”      Cache   Translate Page      

A tecnologia alterou o ambiente de trabalho com a quarta revolução industrial. Atualmente, por exemplo, as máquinas estão dotadas de inteligência e da capacidade de se aperfeiçoar aos diferentes cenários — como o machine learning. Toda essa realidade transformou as relações e as obrigações dos seres humanos, introduzindo o conceito de trabalho 4.0. […]

O post Kienbaum discute sobre o tema “trabalho 4.0” apareceu primeiro em BrasilAlemanha News.


          PyCoder’s Weekly: Issue #341 (Nov. 6, 2018)      Cache   Translate Page      
Come work on PyPI, the future of Python packaging, and more body,#bodyTable,#bodyCell{ height:100% !important; margin:0; padding:0; width:100% !important; } table{ border-collapse:collapse; } img,a img{ border:0; outline:none; text-decoration:none; } h1,h2,h3,h4,h5,h6{ margin:0; padding:0; } p{ margin:1em 0; padding:0; } a{ word-wrap:break-word; } .mcnPreviewText{ display:none !important; } .ReadMsgBody{ width:100%; } .ExternalClass{ width:100%; } .ExternalClass,.ExternalClass p,.ExternalClass span,.ExternalClass font,.ExternalClass td,.ExternalClass div{ line-height:100%; } table,td{ mso-table-lspace:0pt; mso-table-rspace:0pt; } #outlook a{ padding:0; } img{ -ms-interpolation-mode:bicubic; } body,table,td,p,a,li,blockquote{ -ms-text-size-adjust:100%; -webkit-text-size-adjust:100%; } #bodyCell{ padding:0; } .mcnImage,.mcnRetinaImage{ vertical-align:bottom; } .mcnTextContent img{ height:auto !important; } body,#bodyTable{ background-color:#F2F2F2; } #bodyCell{ border-top:0; } h1{ color:#555 !important; display:block; font-family:Helvetica; font-size:40px; font-style:normal; font-weight:bold; line-height:125%; letter-spacing:-1px; margin:0; text-align:left; } h2{ color:#404040 !important; display:block; font-family:Helvetica; font-size:26px; font-style:normal; font-weight:bold; line-height:125%; letter-spacing:-.75px; margin:0; text-align:left; } h3{ color:#555 !important; display:block; font-family:Helvetica; font-size:18px; font-style:normal; font-weight:bold; line-height:125%; letter-spacing:-.5px; margin:0; text-align:left; } h4{ color:#808080 !important; display:block; font-family:Helvetica; font-size:16px; font-style:normal; font-weight:bold; line-height:125%; letter-spacing:normal; margin:0; text-align:left; } #templatePreheader{ background-color:#3399cc; border-top:0; border-bottom:0; } .preheaderContainer .mcnTextContent,.preheaderContainer .mcnTextContent p{ color:#ffffff; font-family:Helvetica; font-size:11px; line-height:125%; text-align:left; } .preheaderContainer .mcnTextContent a{ color:#ffffff; font-weight:normal; text-decoration:underline; } #templateHeader{ background-color:#FFFFFF; border-top:0; border-bottom:0; } .headerContainer .mcnTextContent,.headerContainer .mcnTextContent p{ color:#555; font-family:Helvetica; font-size:15px; line-height:150%; text-align:left; } .headerContainer .mcnTextContent a{ color:#6DC6DD; font-weight:normal; text-decoration:underline; } #templateBody{ background-color:#FFFFFF; border-top:0; border-bottom:0; } .bodyContainer .mcnTextContent,.bodyContainer .mcnTextContent p{ color:#555; font-size:16px; line-height:150%; text-align:left; margin: 0 0 1em 0; } .bodyContainer .mcnTextContent a{ color:#6DC6DD; font-weight:normal; text-decoration:underline; } #templateFooter{ background-color:#F2F2F2; border-top:0; border-bottom:0; } .footerContainer .mcnTextContent,.footerContainer .mcnTextContent p{ color:#555; font-family:Helvetica; font-size:11px; line-height:125%; text-align:left; } .footerContainer .mcnTextContent a{ color:#555; font-weight:normal; text-decoration:underline; } @media only screen and (max-width: 480px){ body,table,td,p,a,li,blockquote{ -webkit-text-size-adjust:none !important; } } @media only screen and (max-width: 480px){ body{ width:100% !important; min-width:100% !important; } } @media only screen and (max-width: 480px){ .mcnRetinaImage{ max-width:100% !important; } } @media only screen and (max-width: 480px){ table[class=mcnTextContentContainer]{ width:100% !important; } } @media only screen and (max-width: 480px){ .mcnBoxedTextContentContainer{ max-width:100% !important; min-width:100% !important; width:100% !important; } } @media only screen and (max-width: 480px){ table[class=mcpreview-image-uploader]{ width:100% !important; display:none !important; } } @media only screen and (max-width: 480px){ img[class=mcnImage]{ width:100% !important; } } @media only screen and (max-width: 480px){ table[class=mcnImageGroupContentContainer]{ width:100% !important; } } @media only screen and (max-width: 480px){ td[class=mcnImageGroupContent]{ padding:9px !important; } } @media only screen and (max-width: 480px){ td[class=mcnImageGroupBlockInner]{ padding-bottom:0 !important; padding-top:0 !important; } } @media only screen and (max-width: 480px){ tbody[class=mcnImageGroupBlockOuter]{ padding-bottom:9px !important; padding-top:9px !important; } } @media only screen and (max-width: 480px){ table[class=mcnCaptionTopContent],table[class=mcnCaptionBottomContent]{ width:100% !important; } } @media only screen and (max-width: 480px){ table[class=mcnCaptionLeftTextContentContainer],table[class=mcnCaptionRightTextContentContainer],table[class=mcnCaptionLeftImageContentContainer],table[class=mcnCaptionRightImageContentContainer],table[class=mcnImageCardLeftTextContentContainer],table[class=mcnImageCardRightTextContentContainer],.mcnImageCardLeftImageContentContainer,.mcnImageCardRightImageContentContainer{ width:100% !important; } } @media only screen and (max-width: 480px){ td[class=mcnImageCardLeftImageContent],td[class=mcnImageCardRightImageContent]{ padding-right:18px !important; padding-left:18px !important; padding-bottom:0 !important; } } @media only screen and (max-width: 480px){ td[class=mcnImageCardBottomImageContent]{ padding-bottom:9px !important; } } @media only screen and (max-width: 480px){ td[class=mcnImageCardTopImageContent]{ padding-top:18px !important; } } @media only screen and (max-width: 480px){ td[class=mcnImageCardLeftImageContent],td[class=mcnImageCardRightImageContent]{ padding-right:18px !important; padding-left:18px !important; padding-bottom:0 !important; } } @media only screen and (max-width: 480px){ td[class=mcnImageCardBottomImageContent]{ padding-bottom:9px !important; } } @media only screen and (max-width: 480px){ td[class=mcnImageCardTopImageContent]{ padding-top:18px !important; } } @media only screen and (max-width: 480px){ table[class=mcnCaptionLeftContentOuter] td[class=mcnTextContent],table[class=mcnCaptionRightContentOuter] td[class=mcnTextContent]{ padding-top:9px !important; } } @media only screen and (max-width: 480px){ td[class=mcnCaptionBlockInner] table[class=mcnCaptionTopContent]:last-child td[class=mcnTextContent],.mcnImageCardTopImageContent,.mcnCaptionBottomContent:last-child .mcnCaptionBottomImageContent{ padding-top:18px !important; } } @media only screen and (max-width: 480px){ td[class=mcnBoxedTextContentColumn]{ padding-left:18px !important; padding-right:18px !important; } } @media only screen and (max-width: 480px){ td[class=mcnTextContent]{ padding-right:18px !important; padding-left:18px !important; } } @media only screen and (max-width: 480px){ table[class=templateContainer]{ max-width:600px !important; width:100% !important; } } @media only screen and (max-width: 480px){ h1{ font-size:24px !important; line-height:125% !important; } } @media only screen and (max-width: 480px){ h2{ font-size:20px !important; line-height:125% !important; } } @media only screen and (max-width: 480px){ h3{ font-size:18px !important; line-height:125% !important; } } @media only screen and (max-width: 480px){ h4{ font-size:16px !important; line-height:125% !important; } } @media only screen and (max-width: 480px){ table[class=mcnBoxedTextContentContainer] td[class=mcnTextContent],td[class=mcnBoxedTextContentContainer] td[class=mcnTextContent] p{ font-size:18px !important; line-height:125% !important; } } @media only screen and (max-width: 480px){ table[id=templatePreheader]{ display:block !important; } } @media only screen and (max-width: 480px){ td[class=preheaderContainer] td[class=mcnTextContent],td[class=preheaderContainer] td[class=mcnTextContent] p{ font-size:14px !important; line-height:115% !important; } } @media only screen and (max-width: 480px){ td[class=headerContainer] td[class=mcnTextContent],td[class=headerContainer] td[class=mcnTextContent] p{ font-size:18px !important; line-height:125% !important; } } @media only screen and (max-width: 480px){ td[class=bodyContainer] td[class=mcnTextContent],td[class=bodyContainer] td[class=mcnTextContent] p{ font-size:18px !important; line-height:125% !important; } } @media only screen and (max-width: 480px){ td[class=footerContainer] td[class=mcnTextContent],td[class=footerContainer] td[class=mcnTextContent] p{ font-size:14px !important; line-height:115% !important; } } @media only screen and (max-width: 480px){ td[class=footerContainer] a[class=utilityLink]{ display:block !important; } }
PSF: Upcoming Contract Work on PyPI
#341 – NOVEMBER 6, 2018 VIEW IN BROWSER
The PyCoder’s Weekly Logo
PSF: Upcoming Contract Work on PyPI
If you have experience with security features or localization features in Python codebases, this is an opportunity to get involved with PyPI. You can register your interest to participate as a contractor online. The project begins in January 2019.
PYTHON SOFTWARE FOUNDATION

The Best Flake8 Extensions for Your Python Project
The flake8 code linter supports plugins that can check for additional rule violations. This post goes into the author’s favorite plugins. I didn’t know flake8-import-order was a thing and I will definitely try this out in my own projects.
JULIEN DANJOU

“Deal With It” Meme GIF Generator Using Python + OpenCV
How to create animated GIFs using OpenCV, Python, and ImageMagick. Super-detailed tutorial and the results are awesome.
ADRIAN ROSEBROCK

Find a Python Job Through Vettery
alt Vettery specializes in developer roles and is completely free for job seekers. Interested? Submit your profile, and if accepted onto the platform, you can receive interview requests directly from top companies seeking Python developers. Get Started.
VETTERYsponsor

Python 2.7 Halloween Facepaint
Scary!
REDDIT.COM

Writing Comments in Python (Guide)
How to write Python comments that are clean, concise, and useful. Get up to speed on what the best practices are, which types of comments it’s best to avoid, and how you can practice writing cleaner comments.
REAL PYTHON

pyproject.toml: The Future of Python Packaging
Deep dive with Brett Cannon into changes to Python packaging such as pyproject.toml, PEP 517, 518, and the implications of these changes. Lots of things happening in that area and this interview is a great way to stay up to date.
TESTANDCODE.COM podcast

Crash Reporting in Desktop Python Applications
The Dropbox desktop client is partly written in Python. This post goes into how their engineering teams do live crash-reporting in their desktop app. Also check out the related slide deck.
DROPBOX.COM


Discussions


When to Use @staticmethod vs Writing a Plain Function?
MAIL.PYTHON.ORG

Can a Non-Python-Programmer Set Up a Django Website With a Few Hours of Practice?
REDDIT.COM

Python Interview Question Post-Mortem
The question was how to merge two lists together in Python (without duplicates.) Interviewers want to see a for-loop solution, even though it’s much slower than what the applicant came up with initially. Good read on what to do/what to avoid if you have a coding interview coming up.
REDDIT.COM

I Just Got a $67k Job Before I Even Graduated, All Thanks to Python
REDDIT.COM


Python Jobs


Senior Software Engineer - Full Stack (Raleigh, North Carolina)
SUGARCRM

Head of Engineering (Remote, Work from Anywhere)
FINDKEEP.LOVE

Senior Developer (Chicago, Illinois)
PANOPTA

Senior Software Engineer (Los Angeles, California)
GOODRX

More Python Jobs >>>


Articles & Tutorials


Setting Up Python for Machine Learning on Windows
In this step-by-step tutorial, you’ll cover the basics of setting up a Python numerical computation environment for machine learning on a Windows machine using the Anaconda Python distribution.
REAL PYTHON

Diving Into Pandas Is Faster Than Reinventing It
How modern Pandas makes your life easier by making your code easier to read—and easier to write.
DEAN LANGSAM • Shared by Dean Langsam

“Ultimate Programmer Super Stack” Bundle [90% off]
Become a well-rounded developer with this book & course bundle. Includes 25+ quality resources for less than $2 each. If you’re looking to round out your reading list for the cold months of the year, this is a great deal. Available this week only.
INFOSTACK.IO sponsor

Structure of a Flask Project
Suggestions for the folder structure of a Flask project. Nice and clean!
LEPTURE.COM • Shared by Python Bytes FM

Dockerizing Django With Postgres, Gunicorn, and Nginx
How to configure Django to run on Docker along with PostgreSQL, Nginx, and Gunicorn.
MICHAEL HERMAN

Making Python Project Executables With PEX
PEX files are distributable Python environments you can use to build executables for your project. These executables can then be copied to the target host and executed there without requiring an install step. This tutorial goes into how to build a PEX file for a simple Click CLI app.
PETER DEMIN

I Was Looking for a House, So I Built a Web Scraper in Python
MEDIUM.COM/@FNEVES • Shared by Ricky White

A Gentle Visual Intro to Data Analysis in Python Using Pandas
Short & sweet intro to basic Pandas concepts. Lots of images and visualizations in there make the article an easy read.
JAY ALAMMAR

Packaging and Developing Python Projects With Nested Git-Submodules
Working with repositories that have nested Git submodules of arbitrary depth, in the context of a Python project. Personally I’m having a hard time working effectively with Git submodules, but if they’re a good fit for your use case check out this article.
KONSTANTINOS DEMARTINOS

Python vs NumPy vs Nim Performance Comparison
Also check out the related discussion on Reddit.
NARIMIRAN.GITHUB.IO

Speeding Up JSON Schema Validation in Python
PETERBE.COM

Careful With Negative Assertions
A cautionary tale about testing that things are unequal…
NED BATCHELDER

Data Manipulation With Pandas: A Brief Tutorial
Covers three basic data manipulation techniques with Pandas: Modifying a DataFrame using the inplace parameter, grouping using groupby(), and handling missing data.
ERIK MARSJA

Full-Stack Developers, Unicorns and Other Mythological Beings
What’s a “Full-Stack” developer anyway?
MEDIUM.COM/DATADRIVENINVESTOR • Shared by Ricky White

Writing Custom Celery Task Loggers
The celery.task logger is used for logging task-specific information, which is useful if you need to know which task a log message came from.
BJOERN STIEL

Generating Software Tests Automatically
An online textbook on automating software testing, specifically by generating tests automatically. Covers random fuzzing, mutation-based fuzzing, grammar-based test generation, symbolic testing, and more. Examples use Python.
FUZZINGBOOK.ORG

Custom User Models in Django
How and why to add a custom user model to your Django project.
WSVINCENT.COM • Shared by Ricky White


Projects & Code


Vespene: Python CI/CD and Automation Server Written in Django
VESPENE.IO

zulu: A Drop-In Replacement for Native Python Datetimes That Embraces UTC
A drop-in replacement for native datetime objects that always uses UTC. Makes it easy to reason about zulu objects. Also conveniently parses ISO8601 and timestamps by default without any extra arguments.
DERRICK GILLAND • Shared by Derrick Gilland

My Python Examples (Scripts)
Little scripts and tools written by someone who says they’re “not a programmer.” Maybe the code quality isn’t perfect here—but hey, if you’re looking for problems to solve with Python, why not do something similar or contribute to this project by improving the scripts?
GITHUB.COM/GEEKCOMPUTERS

termtosvg: Record Terminal Sessions as SVG Animations
A Unix terminal recorder written in Python that renders your command line sessions as standalone SVG animations.
GITHUB.COM/NBEDOS

CPython Speed Center
A performance analysis tool for CPython. It shows performance regressions and allows comparing different applications or implementations over time.
SPEED.PYTHON.ORG

ase: Atomic Simulation Environment
A Python library for working with atoms. There’s a library on PyPI for everything…
GITLAB.COM/ASE

Various Pandas Solutions and Examples
PYTHONPROGRAMMING.IN • Shared by @percy_io

pymc-learn: Probabilistic Models for Machine Learning
Uses a familiar scikit-learn syntax.
PYMC-LEARN.ORG

ReviewNB: Jupyter Notebook Diff for GitHub
HTML-rendered diffs for Jupyter Notebooks. Say goodbye to messy JSON diffs and collaborate on notebooks via review comments.
REVIEWNB.COM


Events


Python LX
14 Nov. in Lisbon, Portugal
PYTHON.ORG

PyData Bristol Meetup (Nov 13)
PYTHON.ORG

Python Miami
10 Nov. – 11 Nov. in Miami, FL.
PYTHON.ORG

Happy Pythoning!
Copyright © 2018 PyCoder’s Weekly, All rights reserved.
alt

[ Subscribe to 🐍 PyCoder’s Weekly 💌 – Get the best Python news, articles, and tutorials delivered to your inbox once a week >> Click here to learn more ]


          Stack Abuse: Applying Wrapper Methods in Python for Feature Selection      Cache   Translate Page      

Introduction

In the previous article, we studied how we can use filter methods for feature selection for machine learning algorithms. Filter methods are handy when you want to select a generic set of features for all the machine learning models.

However, in some scenarios, you may want to use a specific machine learning algorithm to train your model. In such cases, features selected through filter methods may not be the most optimal set of features for that specific algorithm. There is another category of feature selection methods that select the most optimal features for the specified algorithm. Such methods are called wrapper methods.

Wrapper Methods for Feature Selection

Wrapper methods are based on greedy search algorithms as they evaluate all possible combinations of the features and select the combination that produces the best result for a specific machine learning algorithm. A downside to this approach is that testing all possible combinations of the features can be computationally very expensive, particularly if the feature set is very large.

As said earlier, wrapper methods can find the best set of features for a specific algorithm - however, a downside is that these set of features may not be optimal for every other machine learning algorithm.

Wrapper methods for feature selection can be divided into three categories: Step forward feature selection, Step backwards feature selection and Exhaustive feature selection. In this article, we will see how we can implement these feature selection approaches in Python.

Step Forward Feature Selection

In the first phase of the step forward feature selection, the performance of the classifier is evaluated with respect to each feature. The feature that performs the best is selected out of all the features.

In the second step, the first feature is tried in combination with all the other features. The combination of two features that yield the best algorithm performance is selected. The process continues until the specified number of features are selected.

Let's implement step forward feature selection in Python. We will be using the BNP Paribas Cardif Claims Management dataset for this section as we did in our previous article.

To implement step forward feature selection, we need to convert categorical feature values into numeric feature values. However, for the sake of simplicity, we will remove all the non-categorical columns from our data. We will also remove the correlated columns as we did in the previous article so that we have a small feature set to process.

Data Preprocessing

The following script imports the dataset and the required libraries, it then removes the non-numeric columns from the dataset and then divides the dataset into training and testing sets. Finally, all the columns with a correlation of greater than 0.8 are removed. Take a look at this article for the detailed explanation of this script:

import pandas as pd  
import numpy as np  
from sklearn.model_selection import train_test_split  
from sklearn.feature_selection import VarianceThreshold

paribas_data = pd.read_csv(r"E:\Datasets\paribas_data.csv", nrows=20000)  
paribas_data.shape

num_colums = ['int16', 'int32', 'int64', 'float16', 'float32', 'float64']  
numerical_columns = list(paribas_data.select_dtypes(include=num_colums).columns)  
paribas_data = paribas_data[numerical_columns]  
paribas_data.shape

train_features, test_features, train_labels, test_labels = train_test_split(  
    paribas_data.drop(labels=['target', 'ID'], axis=1),
    paribas_data['target'],
    test_size=0.2,
    random_state=41)

correlated_features = set()  
correlation_matrix = paribas_data.corr()  
for i in range(len(correlation_matrix .columns)):  
    for j in range(i):
        if abs(correlation_matrix.iloc[i, j]) > 0.8:
            colname = correlation_matrix.columns[i]
            correlated_features.add(colname)


train_features.drop(labels=correlated_features, axis=1, inplace=True)  
test_features.drop(labels=correlated_features, axis=1, inplace=True)

train_features.shape, test_features.shape  
Implementing Step Forward Feature Selection in Python

To select the most optimal features, we will be using SequentialFeatureSelector function from the mlxtend library. The library can be downloaded executing the following command at anaconda command prompt:

conda install -c conda-forge mlxtend  

We will use the Random Forest Classifier to find the most optimal parameters. The evaluation criteria used will be ROC-AUC. The following script selects the 15 features from our dataset that yields best performance for random forest classifier:

from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier  
from sklearn.metrics import roc_auc_score

from mlxtend.feature_selection import SequentialFeatureSelector

feature_selector = SequentialFeatureSelector(RandomForestClassifier(n_jobs=-1),  
           k_features=15,
           forward=True,
           verbose=2,
           scoring='roc_auc',
           cv=4)

In the script above we pass the RandomForestClassifieras the estimator to the SequentialFeatureSelector function. The k_features specifies the number of features to select. You can set any number of features here. The forward parameter, if set to True, performs step forward feature selection. The verbose parameter is used for logging the progress of the feature selector, the scoring parameter defines the performance evaluation criteria and finally, cv refers to cross-validation folds.

We created our feature selector, now we need to call the fit method on our feature selector and pass it the training and test sets as shown below:

features = feature_selector.fit(np.array(train_features.fillna(0)), train_labels)  

Depending upon your system hardware, the above script can take some time to execute. Once the above script finishes executing, you can execute the following script to see the 15 selected features:

filtered_features= train_features.columns[list(features.k_feature_idx_)]  
filtered_features  

In the output, you should see the following features:

Index(['v4', 'v10', 'v14', 'v15', 'v18', 'v20', 'v23', 'v34', 'v38', 'v42',  
       'v50', 'v51', 'v69', 'v72', 'v129'],
      dtype='object')

Now to see the classification performance of the random forest algorithm using these 15 features, execute the following script:

clf = RandomForestClassifier(n_estimators=100, random_state=41, max_depth=3)  
clf.fit(train_features[filtered_features].fillna(0), train_labels)

train_pred = clf.predict_proba(train_features[filtered_features].fillna(0))  
print('Accuracy on training set: {}'.format(roc_auc_score(train_labels, train_pred[:,1])))

test_pred = clf.predict_proba(test_features[filtered_features].fillna(0))  
print('Accuracy on test set: {}'.format(roc_auc_score(test_labels, test_pred [:,1])))  

In the script above, we train our random forest algorithm on the 15 features that we selected using the step forward feature selection and then we evaluated the performance of our algorithm on the training and testing sets. In the output, you should see the following results:

Accuracy on training set: 0.7072327148174093  
Accuracy on test set: 0.7096973252804142  

You can see that the accuracy on training and test sets is pretty similar which means that our model is not overfitting.

Step Backwards Feature Selection

Step backwards feature selection, as the name suggests is the exact opposite of step forward feature selection that we studied in the last section. In the first step of the step backwards feature selection, one feature is removed in round-robin fashion from the feature set and the performance of the classifier is evaluated.

The feature set that yields the best performance is retained. In the second step, again one feature is removed in a round-robin fashion and the performance of all the combination of features except the 2 features is evaluated. This process continues until the specified number of features remain in the dataset.

Step Backwards Feature Selection in Python

In this section, we will implement the step backwards feature selection on the BNP Paribas Cardif Claims Management. The preprocessing step will remain the same as the previous section. The only change will be in the forward parameter of the SequentiaFeatureSelector class. In case of the step backwards feature selection, we will set this parameter to False. Execute the following script:

from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier  
from sklearn.metrics import roc_auc_score  
from mlxtend.feature_selection import SequentialFeatureSelector

feature_selector = SequentialFeatureSelector(RandomForestClassifier(n_jobs=-1),  
           k_features=15,
           forward=False,
           verbose=2,
           scoring='roc_auc',
           cv=4)

features = feature_selector.fit(np.array(train_features.fillna(0)), train_labels)  

To see the feature selected as a result of step backwards elimination, execute the following script:

filtered_features= train_features.columns[list(features.k_feature_idx_)]  
filtered_features  

The output looks like this:

Index(['v7', 'v8', 'v10', 'v17', 'v34', 'v38', 'v45', 'v50', 'v51', 'v61',  
       'v94', 'v99', 'v119', 'v120', 'v129'],
      dtype='object')

Finally, let's evaluate the performance of our random forest classifier on the features selected as a result of step backwards feature selection. Execute the following script:

clf = RandomForestClassifier(n_estimators=100, random_state=41, max_depth=3)  
clf.fit(train_features[filtered_features].fillna(0), train_labels)

train_pred = clf.predict_proba(train_features[filtered_features].fillna(0))  
print('Accuracy on training set: {}'.format(roc_auc_score(train_labels, train_pred[:,1])))

test_pred = clf.predict_proba(test_features[filtered_features].fillna(0))  
print('Accuracy on test set: {}'.format(roc_auc_score(test_labels, test_pred [:,1])))  

The output looks likes that:

Accuracy on training set: 0.7095207938140247  
Accuracy on test set: 0.7114624676445211  

You can see that the performance achieved on the training set is similar to that achieved using step forward feature selection. However, on the test set, backward feature selection performed slightly better.

Exhaustive Feature Selection

In exhaustive feature selection, the performance of a machine learning algorithm is evaluated against all possible combinations of the features in the dataset. The feature subset that yields best performance is selected. The exhaustive search algorithm is the most greedy algorithm of all the wrapper methods since it tries all the combination of features and selects the best.

A downside to exhaustive feature selection is that it can be slower compared to step forward and step backward method since it evaluates all feature combinations.

Exhaustive Feature Selection in Python

In this section, we will implement the step backwards feature selection on the BNP Paribas Cardif Claims Management. The preprocessing step will remain the similar to that of Step forward feature selection.

To implement exhaustive feature selection, we will be using ExhaustiveFeatureSelector function from the mlxtend.feature_selection library. The class has min_featuresand max_features attributes which can be used to specify the minimum and the maximum number of features in the combination.

Execute the following script:

from mlxtend.feature_selection import ExhaustiveFeatureSelector  
from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier  
from sklearn.metrics import roc_auc_score

feature_selector = ExhaustiveFeatureSelector(RandomForestClassifier(n_jobs=-1),  
           min_features=2,
           max_features=4,
           scoring='roc_auc',
           print_progress=True,
           cv=2)

We created our feature selector, now need to call the fit method on our feature selector and pass it the training and test sets as shown below:

features = feature_selector.fit(np.array(train_features.fillna(0)), train_labels)  

Note that the above script can take quite a bit of time to execute. To see the feature selected as a result of step backwards elimination, execute the following script:

filtered_features= train_features.columns[list(features.k_feature_idx_)]  
filtered_features  

Finally, to see the performance of random forest classifier on the features selected as a result of exhaustive feature selection. Execute the following script:

clf = RandomForestClassifier(n_estimators=100, random_state=41, max_depth=3)  
clf.fit(train_features[filtered_features].fillna(0), train_labels)

train_pred = clf.predict_proba(train_features[filtered_features].fillna(0))  
print('Accuracy on training set: {}'.format(roc_auc_score(train_labels, train_pred[:,1])))

test_pred = clf.predict_proba(test_features[filtered_features].fillna(0))  
print('Accuracy on test set: {}'.format(roc_auc_score(test_labels, test_pred [:,1])))  

Conclusion

Wrapper methods are some of the most important algorithms used for feature selection for a specific machine learning algorithm. In this article, we studied different types of wrapper methods along with their practical implementation. We studied step forward, step backwards and exhaustive methods for feature selection.

As a rule of thumb, if the dataset is small, exhaustive feature selection method should be the choice, however, in case of large datasets, step forward or step backward feature selection methods should be preferred.


          On a Podcast      Cache   Translate Page      
I’m on https://www.datacamp.com/community/podcast/human-centered-design-data-science at 38:10 talking about interpretability and fairness in Machine Learning. Advertisements
          Matrox Imaging: Flowchart software      Cache   Translate Page      
Matrox® Imaging announces Matrox Design Assistant X flowchart-based vision application software. This integrated development environment (IDE) allows developers to build intuitive flowcharts instead of writing traditional program code. It enables the development of a graphical web-based operator interface for modifying the vision application.

This update integrates a host of new features and functionality, including image classification using deep learning, a photometric stereo tool that highlight surface imperfections, and the ability to interface directly with third-party 3D sensors.

Deep learning for image classification

The classification tool leverages deep learning—specifically, convolutional neural network (CNN) technology—to categorize images of highly textured, naturally varying, and acceptably deformed goods. All inference is performed on a mainstream CPU, eliminating the dependence on third-party neural network libraries and the need for specialized GPU hardware. Matrox Imaging handles the intricate design and training of the neural network, utilizing the deep technical experience, knowledge, and skill of its machine learning and machine vision experts.

A Q&A video offers more insight into deep learning technology.

Photometric stereo for emphasizing surface irregularities

A new registration tool features photometric stereo technology, which creates a composite image from a series of images taken with light coming in from different directions. Creation of these images utilizes directional illumination light controllers, such as the Light Sequence Switch (LSS) from CCS, LED Light Manager (LLM) from Smart Vision Lights, or others similar. This composite image emphasizes surface irregularities, such as embossed or engraved features, scratches, or indentations.

A primer on photometric stereo techniques was outlined in a Q&A video.

Third-party 3D sensor interfacing

Matrox Design Assistant X makes it possible to capture and process depth-map data byinterfacing with third-party 3D sensors. Initially, the software will support LMI Gocator® line profilers and snapshot sensors and Photoneo® PhoXi® scanners, with other scanner options to be added in the future.

Other updates and additions include multiple run-times for running multiple independent projects simultaneously on the same platform; dedicated shape-finding tools for locating circles, ellipses, rectangles, and line segments; and addition of a code-grading step.

Field-proven Matrox Design Assistant X software is a perfect match for the Matrox 4Sight EV6 vision controller or the Matrox Iris GTR smart camera.

"This new version delivers on the three cornerstones of our development methodology," said Fabio Perelli, product manager, Matrox Imaging. "These are to extend Matrox Design Assistant’s capabilities while incorporating recent evolutions to the underlying vision library and also striving to simplify the overall user experience."

Availability
Matrox Design Assistant X will be officially released in Q2 2019.


          Data scientist      Cache   Translate Page      
Looking for someone to contribute to an ongoing project as data analyst. You are expected to be having experience on machine learning and deep learning modules using Python. Minimum 4 hours a days is needed to be spent on the project... (Budget: ₹100 - ₹400 INR, Jobs: Data Mining, Machine Learning, Python, Software Architecture, Statistics)
          Consultant, Business Analytics & Data Science - Lincoln Financial - Boston, MA      Cache   Translate Page      
Phoenix, AZ (Arizona). Knowledge and experience on applying statistical and machine learning techniques on real business data....
From Lincoln Financial Group - Fri, 02 Nov 2018 02:54:18 GMT - View all Boston, MA jobs
          AWS Architect - Insight Enterprises, Inc. - Chicago, IL      Cache   Translate Page      
Database architecture, Big Data, Machine Learning, Business Intelligence, Advanced Analytics, Data Mining, ETL. Internal teammate application guidelines:....
From Insight - Thu, 12 Jul 2018 01:56:10 GMT - View all Chicago, IL jobs
          How a Windows Update Can Spark New IoT Services on Campuses      Cache   Translate Page      
How a Windows Update Can Spark New IoT Services on Campuses eli.zimmerman_9856 Tue, 11/06/2018 - 13:29

Microsoft announced a new update to its Windows 10 OS in October that will open doors to a new level of connectedness on college campuses.

Integration of the Internet of Things has become a common goal for higher education institutions as universities strive for a connected campus model. 

While many colleges are focusing heavily on IoT as part of their digital transformation initiatives, administrators and IT leaders are facing certain hurdles, particularly concerning data quality and management, which has slowed down campus innovation.

"Data management is the biggest obstacle we have right now," Gerry Hamilton, director of facilities energy management at Stanford University tells Campus Technology. "It all comes down to scalability and sustainability. We have found there is an exponential growth of effort that happens every time you deploy one more system."

The new update promises to bring “edge intelligence with machine learning, industrial strength security” and “diverse silicon options” to users. 

MORE FROM EDTECH: Check out how universities are preparing their networks for IoT integration!

Machine Learning Improves IoT Management

To help users get a firm grasp on their IoT data management, the new Windows 10 update will employ Microsoft Azure IoT Edge and Windows machine learning.

“Windows 10 IoT enables you to do more at the edge including machine learning, event processing, image recognition and other high-value artificial intelligence without developing it in-house,” according to Microsoft’s announcement. “Seamless integration with Azure IoT Edge brings cloud intelligence and analytics securely to Windows 10 IoT devices at scale.”

Azure’s IoT Edge service will allow campus IT teams to run AI workloads and Azure services on Windows 10 IoT devices locally and remotely, easing the burden on campuses to create a web of connected devices. 

At the University of Nebraska–Lincoln, campus staff members are looking into using Azure IoT Edge to help manage sensors that detect facility issues on campus, EdTech reports

“Cloud-based machine learning applications in Microsoft Azure and similar technologies may help systems learn how to reduce the time needed to identify HVAC faults,” says Lalit Agarwal, the university’s director of utility and energy management. “That is definitely an area for us to investigate for the future.”

In addition, the integration of Azure IoT Device Management and Microsoft Intune simplifies device monitoring, allowing campus IT teams to develop management solutions and consolidate device management across a single interface.

Integrate and Analyze Data from Campus Points of Sale

Digital kiosks have become key additions to campus stores, restaurants and stadiums in support of universities’ digital transformation agendas. 

At Clemson University, kiosks in campus mailrooms have helped cut wait times for packages down from 40 minutes to an average of one minute

Through the Windows 10 update, IT teams will find it easier to customize and manage kiosks on campus. Through assigned access, managers can “customize the functionality exposed by kiosks and other fixed-function devices, providing a streamlined, intuitive user experience that is focused on specific tasks,” according to Microsoft.

The update will also provide enhanced status reporting, automatically alerting IT teams when a kiosk is experiencing problems, as well as initiating corrective responses like restarting the device.

Digital%20Transformation_IR_1.jpg

Eli has been eagerly pursuing a journalistic career since he left the University of Maryland's Philip Merrill School of Journalism. Previously, Eli was a staff reporter for medical trade publication Frontline Medical News, where he experienced the impact of continuous education and evolving teaching methods through the medical lens. When not in the office, Eli is busy scanning the web for the latest podcasts or stepping into the boxing ring for a few rounds.


          Improving the GoDaddy User Experience with Elastic Machine Learning      Cache   Translate Page      

This post is a recap of a community talk given at Elastic{ON} 2018. Interested in seeing more talks like this? Check out the conference archive or find out when the Elastic{ON} Tour is coming to a city near you.

GoDaddy is known for web hosting and domain management, as anyone that’s watched the Super Bowl in recent years would know. But with over 17 million customers, 75 million domains, and 10 million hosted sites, they’re also well versed in big data. Keeping sites running smoothly requires insight into every piece of their infrastructure, from virtual server patch level to network hiccups to malicious attacks. This could be difficult with over 200,000 messages coming in every second (DNS queries, system logs, business events, and more), but with its speed at scale, the Elastic Stack is up to the task.

GoDaddy’s introduction to Elasticsearch was a lot like other companies that use open source software. Disparate teams throughout the company set up their own clusters to handle their own specific needs. It got the job done, but this unmanaged deployment model led to hundreds of clusters running on varying versions of Elasticsearch analyzing siloed data. Knowing there was a better way, they formed a team around managing the deployment of Elasticsearch in 2014. This team now manages over 60 Elasticsearch clusters spanning 700+ Docker containers, with feeds coming in from teams all over the company. These clusters account for over 270 TB of data from their (11 PB) HDFS environment.

One of the first use cases their new Elasticsearch team tackled was managing patch compliance throughout their entire ecosystem. In the pre-Beats world of 2014, GoDaddy developed Windows and Linux agents (similar to Auditbeat and Winlogbeat) to send system data to Elasticsearch. With these agents installed on all of their servers (bare metal and virtual), GoDaddy was able to gain valuable insight into patching levels and compliance throughout their entire infrastructure. And by utilizing different dashboards and visualizations within Kibana, they were able to easily provide fine-grain patch information to admins and engineers, as well as high-level business reports to executives — all while accessing the same centralized data so everyone is on the same page.

Maintaining server patch levels is important for keeping site traffic flowing, and that flow helps keep users engaged. If a website is loading slowly, visitors will go somewhere else. So, with the experience of their millions of customers in mind, GoDaddy knew they needed to track how well data centers were performing and how their performance impacted visitors. They already had all of the data they needed, as every component of their systems generated logs, but they needed a way to view it holistically.

Centralized Logging with Machine Learning for Anomaly Detection

GoDaddy needed to centralize and analyze their various performance and engagement data sets, and the Elastic Stack was the answer. By sending netflow data, sFlow data, real user management (RUM), and peering relationship and routing data to Elasticsearch, they were able to get a much more detailed view of user experience and system performance data — a level of detail that can only be seen by analyzing all of the different data sources at once. And since then, GoDaddy has begun to take that data even further with the help of Elastic machine learning features.

Having centralized access to mountains of system data is great, but tracking down problems can be difficult. GoDaddy tracks every user click and website interaction, but with millions of pages operating around the world, there’s no way any team of humans could sift through all that data. Fortunately, Elastic machine learning features make anomaly detection a simple task. Working with machine learning experts at Elastic, GoDaddy has been able to implement RUM-focused machine learning jobs that have made anomaly detection easy.

“In terms of the overall effort, leverage your Elastic team. They are extremely helpful. We've had a very close partnership and very frequent calls, a completely open line of communication around all the updates, you will get stuck, use them for that. That's really what they're good at.” - Felix Gorodishter, Principal Architect, GoDaddy

By specifying a threshold for page load times and parameters around page traffic, the GoDaddy team lets Elastic machine learning features handle the job of learning what’s normal and what’s anomalous, and then letting them know whenever there’s a problem. Machine learning cuts through the noise so GoDaddy can focus on what’s important.

Learn about how GoDaddy is leveraging Elastic machine learning features to monitor hosted site performance by watching Stories from the Trenches at GoDaddy: How Big Data Insights Equal Big Money from Elastic{ON} 2018. You’ll also get a peek into the interesting ways they’re using machine learning to monitor business KPIs around product adoptions and hear about the lessons they’ve learned along the way.


          Imaging and Machine Learning Research Scientist - AIRY:3D - Montréal, QC      Cache   Translate Page      
Contribute to expanding our patent portfolio, generate publishable research. We’re looking to add even more imaging and machine learning expertise to our team....
From Indeed - Wed, 17 Oct 2018 15:54:14 GMT - View all Montréal, QC jobs
          Junior Machine Learning Developer - Gartner - Québec City, QC      Cache   Translate Page      
What makes Gartner a GREAT fit for you? When you join Gartner, you’ll be part of a fast-growing team that helps the world become smarter and more connected. We...
From Gartner, Inc. - Thu, 02 Aug 2018 14:12:39 GMT - View all Québec City, QC jobs
          IBM Watson helps Goldcorp mine its data for value [Ingenious winner]      Cache   Translate Page      
IBM and Goldcorp are using AI and machine learning to squeeze out every bit of gold from an Ontario mine.
          Is 2019 the year the robots take over? Not quite      Cache   Translate Page      

With all the buzz around AI and machine learning, it can seem impossible for many marketers to separate the signal from the noise. Is it the best thing since the internet? Are we all out of a job? As with most new technologies, the truth is slightly less dramatic (at least in the short term), and it helps to be armed with some basic knowledge to cut through the noise.

In reality, these technologies are more and more part of our everyday lives and, most vividly apparent where brands connect with consumers, from Amazon’s Alexa to online chat support. These are just two customer touch points that marketers should be aware of the impact of AI. There are many more, ranging from how and when your ads show up online, to revealing new customer segments that you might market to, or even develop new products for.

So, as we increasingly see this new technology insert itself into new parts of our lives, these are my top truths and myths to help marketers identify what’s worth paying attention to, and what’s just hype.

Three truths about AI

1. Anything that can be automated will be (and probably should be)

Marketers are drowning in data. To be effective, modern marketing needs billions of decisions to be made quickly and accurately every day. These decisions help determine which audiences see which messages and the right level of a client’s budget to assign. Humans cannot be expected to handle all of these decisions and thus we turn to machines. Machines excel at processing vast amounts of data and can analyse complex patterns at scale in milliseconds. What’s more, they’re excellent at identifying patterns that humans might miss, which helps identify emerging areas of demand that a brand may have previously overlooked.

2. When you say AI you probably mean machine learning

While both terms are synonymous for most people, knowing the difference will earn you serious points with any engineers you might work with. AI is a poorly defined term, referring to a field of research concerned with the creation of human-like intelligence. Machine learning is concerned with getting machines to complete tasks without being explicitly programmed to handle those tasks. These machines thus “learn” and get better at subsequent tasks with experience. An example of this would be Amazon’s Alexa. Alexa is not explicitly programmed to recognize all speech, rather the machine learning algorithms that it employs get better at understanding commands by improving its own recognition of different accents and common combinations through exposure to more and more examples.

3. Brands that make better use of first-party data will win

Most of us are familiar with the phrase ‘garbage in, garbage out’. The same holds true for machine learning, but more so. If we, as humans, are capable of making disastrously poor decisions based on incomplete data, imagine that scaled up to the speed and power of modern computing technology.

The biggest culprit of bad machine-led decisions - and cause of executive disappointment - is the use of a hodgepodge of second or third-hand data bought or borrowed from various partners by companies. Invariably out of date and massaged to fit the task at hand, what started as a partial look at a particular audience suddenly risks becoming wildly misleading ‘insights’ that could spell disaster for marketers that bet their careers on them. This sort of data manipulation should be avoided. Instead, marketers should make it their focus to turn over every rock in the search for owned data that is directly linked to customer buying decisions, for example, customer relationship (CRM) systems. These are more complete and up-to-date and will drive better business decisions.

Three myths about AI:

1. The robots are about to take over

While some might be concerned about robot dogs opening doors, (perhaps unnerving for some), we’re still some way off ceding power to our robot overlords. Guru Banavar, the head of the team at IBM responsible for creating Watson (the AI system that mastered Jeopardy) told Tech Republic that most people don't have a good understanding of what machine learning is. He said: "We can teach a computer to recognise a car, but we can't ask that same computer, 'How many wheels does that car have?' or, 'What kind of engine does it have?'. Can you ask anything else about what this car is made of or how it is made? None of those things is possible. Those are all far away."

The type of AI known as Artificial General Intelligence is still a long way off.

2. Machines will take all of our jobs

Not all, but they will do better at some than we ever can. Mundane, repetitive and dangerous jobs will be replaced. This is nothing new and has proven to be a good thing in the past when it comes to our general well-being and longevity, as has been the case as we’ve moved from the fields to the factories to the modern internet age.

The good news is that as some jobs are phased out, as repetitive and rote jobs are automated, new industries will emerge, which will create a host of new opportunities.

3. Machines will swallow all your personal data

It’s true that AI systems have the ability to collect, store and analyse vast amounts of consumer data – but so do non-AI systems. It’s important to note that the same privacy by design principles that should be applied to a traditional approach can and should be applied to AI-based systems (especially in light of the recent GDPR implementation in Europe and new regulations on the horizon in the US and around the world). Consumer data must be handled carefully and transparently. What’s more, companies need to ensure they’re gathering only the minimum amount of data they need to deliver the service consumers expect.

The good news for APAC marketers is that they are ahead of the global curve when it comes to adopting, understanding and embracing the benefits of AI, according to Adobe and Econsultancy’s Digital Intelligence Briefing: 2018 Digital Trends report. Those that are combining digital marketing skills with technology are nearly twice as likely to have surpassed their 2017 business goals by a significant margin (20% vs. 11%).

As data gets more abundant and processing get smarter, the innovation available to marketers will multiply. We’re already seeing it in the way in which consumers engage with brands today, but we can expect it to disrupt every consumer engagement, every company and every industry in the coming years. Armed with these myths and truths, you’ll be better equipped to spot the most trustworthy signals.

Konrad Feldman is founder and CEO of Quantcast.


          Vox pop: When was the year of mobile?      Cache   Translate Page      

To kick off this issue of The Drum magazine dedicated to modern mobile marketing, we check in with mobile experts from around the world for their take on whether we ever did actually have a definitive ‘year of mobile’.

Regina Goh, APAC managing director, Blis: It depends on what perspective we are looking at. From the consumer standpoint, I believe the ‘year of mobile’ happened at least a decade ago when touchscreen phones were made possible at a scalable and accessible level across the world, thanks to Apple’s launch of the iPhone in 2007 and HTC’s first Android phone hitting the market in 2008.

From the advertisers’ point of view, it was definitely an annual affair for industry folks around south-east Asia to gather and often ask ‘is this is the year of mobile, yet?’. I believe the evolution took a slow pace over the last 10 years from advertisers asking, ‘what about mobile?!’ to ‘why mobile?’ to today’s prevailing demand of ‘how, and how much?’ to embrace mobile. Based on our experience over the last 18 months, and with mobile ad spend surpassing desktop for the first time ever, I believe the recognition of mobile finally happened in early 2017. Now that we’re finally here, it’s no longer just a ‘how?’ With the enablement of programmatic and cross-device technology, it’s ‘how to do it properly?’

The spin-off is that mobile is being recognized beyond just a platform. It is a full-fledged discourse on mobility, data and enabling programmatic to leverage the real potential of the platform. 

Guillaume Larrieu, EMEA head of mobile, ad platforms, Oath: I’d say 2013, the first year smartphone sales finally surpassed feature phone sales. That brought tremendous mobile branding experiences at a truly global scale for advertisers.

Gavin Stirrat, consultant, GJS Media: The year of mobile has sadly come and gone. We were probably at peak mobile around 2014, but the shifts in spend to programmatic, and a lack of mobile strategies in programmatic from most big brands, mean that the majority of advertising we see on mobile is performance. There are exceptions to this, and there are decent amounts of mobile-specific spending through ‘managed buys’, but in an ever increasingly programmatic world, the challenges of making the most of mobile specific opportunities has fallen lower down the list of priorities as brands focus much more effort on managing brand safety, limiting fraud and managing costs. 

Ilicco Elia, head of mobile, Deloitte Digital: This may be an obvious answer, but we have not yet hit the ‘year of mobile’.

I first heard the phrase back when I was delivering news via AvantGo to Palm V PDAs in 1999. We believed back then that we had access to, and it was marketed as, a ‘seamless sideloading experience’, which now obviously looks like an oxymoron.

We have yet to even scratch the surface of the influence a supercomputer in your pocket will have on civilization. As experiences continue to improve, the word mobile will become synonymous with ‘personal’ and it will influence everything and everyone you interact with.

The ‘year of mobile’ will be the first year in which your ‘phone’ makes a decision on your behalf with your best interests at heart.

Jack Withey, digital channels performance manager, Barclaycard: 2018 will be the ‘year of mobile’ in general, until 2019 comes around. Every year since 2012 has been the new ‘year of mobile’. The leaps we have seen every year since 2012 lead me to believe this will be the case for many years to come. Mobile video, augmented reality and improvements in multi-channel attribution will be the big areas that will truly affect the customer’s interaction with mobile this year, and I’m sure more groundbreaking technology will come into play in the coming years.

The same can be said for search. Since mobile searches overtook desktop searches in 2015 we have seen some massive improvements in mobile, with Google introducing accelerated mobile pages. And the fact that 2018 will see the implementation of the mobile first index alongside improvements in AI and machine learning for mobile, shows the rise of mobile is not going to slow down.

James Hilton, global chief executive, M&C Saatchi Mobile: The rise, and subsequent dominance, of mobile has been signposted for some time and it was always a case of when, not if, mobile would overtake desktop. Arguably, there has been no fixed ‘year of mobile’. Rather, this is the ‘decade of mobile’.

The big shift that must first happen in 2018 is that mobile should no longer be viewed as a channel and instead be treated as the default view for all digital. Mobile-first has been a media buzz phrase for years now, however it is still far from the de facto approach for most brands.

Mobile-first simply must be the way that all digital channels are approached. Search, display, video, e-commerce, CRM performance, web: everything digital should be planned, designed and executed with the mobile consumer front of mind – a necessity now more than ever.

Jeff Ratner, global chief media officer, iCrossing PR: The ‘year of mobile’ came and went without much fanfare. While the industry had been waiting for this milestone since 2010, when the turning point came, it was barely acknowledged. While Geoff Ramsey of eMarketer may have trumpeted spending in excess of $100m in 2016 from high up on Mt Media, it was drowned out by new planning and buying strategies. We moved from buying defined channels to datainfused audience buying and screen agnostic approaches. As consumers moved to best screen available fluidity for content consumption and time spent so did ad spend.

As eMarketer projects in 2018 that mobile will account for 63.3% of digital and 24.3% of total media ad spending, it is less likely to be bought directly as ‘mobile’ and more likely programmatic spend that is delivered through mobile channels. While there are certainly mobile-only campaigns, and, for specific efforts, plans will over-index mobile spend, the ‘year of mobile’ was quickly replaced by the ‘year of audience over channel’.

Siddarth Correya, APAC managing director, SelectMedia: Among the numerous curiosities of digital advertising, the one that refuses to go away is the decade-long ritual of dubbing every year the ‘year of mobile’. So, have we hit a peak in mobile?

Let’s break this into two threads. Advertisers want eyeballs and reach as fundamental when they allocate investments. The most ubiquitous digital media platforms today are primarily accessed via mobile, and in countries like Indonesia via mobile exclusively. So that media plan comprising search and social is already a ‘mobile plan’.

There is an opportunity to create mobile-only campaigns on platforms which are singularly accessed via mobile – chat and messaging, mobile games and maps to name a few. However, this requires a significant shift from the ‘lets slap a few banners on and see what happens’ approach. We need to shift the conversation from the year of mobile to what we earn from mobile. That is, cleaner data, better content designed for mobile and measurement to gauge the efficacy of the medium.

Richard Downey, global director, mobile, The Specialist Works: I’ve given this question a fair bit of thought over the past year or so and the only answer I am comfortable giving is that every year since 2007 has been the ‘year of mobile’. Of course, in the marketing world there are milestones such as Google buying a tiny Palo Alto start up called Android in 2007, or when Facebook introduced its app install ads in 2012, transforming the mobile app advertising landscape forever. Or maybe the ‘year of mobile’ was 2016, when mobile web usage overtook desktop web usage for the first time. But if we remove ourselves from our marketing bubble for a second and look at the transformative effect mobile has had on the planet as a whole, then other, more important milestones emerge. India notched up its billionth mobile phone subscriber in 2016, and many millions of these subscribers live in rural communities that had never before had a telephone of any kind. So the ‘year of mobile’ is all of those mentioned and no doubt will be all of the forthcoming years as well.

Jo Coombs, chief executive officer, OgilvyOne UK: It’s impossible to choose a single year as the pinnacle for mobile when it’s undoubtedly been an entire decade of smartphone dominance. 2007 changed the way we connect forever with the launch of the first iPhone. What followed was the evolution of social media and apps, all of which provided an opportunity for individuals to connect, share and explore at a pace and depth we couldn’t have foreseen. For brands and marketers, it’s provided an ever-changing landscape to create entertainment, channels to converse and evidently shape modern day life with the quest for convenience constant. Fast forward to today and we are at the height of smartphone penetration, with what is ostensibly a global duopoly between Apple and Android services. The next couple of years could be the most exciting yet as we see that battle play out, voice becomes more dominant and AR finally becomes mainstream.

Gela Fridman, managing director, technology, Huge: 2014 was the year that mobile devices peaked and plateaued in terms of basic form and functionality, app store maturity and user expectations. But 2020 will be the real ‘year of mobile’ when 5G catalyzes the next generation of mobile experiences, enabling seamless and direct interaction between brands and consumers and changing the definition of mobile from what happens on a small screen in your hand to what the device in your pocket activates and enables in the world around you on your behalf.

Ryan Hall, managing director, creative product, Karmarama: Mobile phones have quickly been taken for granted as much as electricity or central heating. We really don’t remember quite what life was like before they existed. And no one expected it to be like this. As the technology and network infrastructure have evolved, the mobile phone has been on a journey. And, arguably, can we definitively say that the ‘year of mobile’ has happened?

When mobile phones were introduced they were viewed as an exclusive form of telephone service that might possibly suit certain mobile workforces, such as craftsmen, photographers and repairmen. The Motorola DynaTAC 8000X was launched in 1983 to the tune of $3,995 and could hold 30 contacts.

When Steve Jobs showed the Apple iPhone to the world, the whole game changed. With its multi-touch screen and rich, interactive, gestural apps, followed by the launch of the App Store only two years later, this device has had an incredible impact on many levels.

The internet of things has added a further dimension to the future of mobile. It will be one of hundreds of connected smart devices in the home by the year 2025. And as the journey of mobile continues, I believe the best is yet to come.

Karen Boswell, head of innovations, Adam & Eve DDB :1998, the year Snake was first preloaded on to Nokia phones. Obvs.

OK, seriously? The ‘year of mobile’ was 2014 because that was the year that mobile inarguably became the ubiquitous extension of the human arm: the term ‘smartphone zombie’ was first established (China even constructed a dedicated smartphonesidewalk), ‘tech neck’ spiked, social media led to a surge of streaming on mobile that overtook desktop by almost a factor of two and smartphone users overtook desktop for the first time.

This feature first appeared in The Drum's March issue, which focused on mobile technology.


          Stage: Afstudeeropdracht richting AI, Machine Learning in Veenendaal      Cache   Translate Page      
<p align="LEFT"><strong>Tracken van gewassen met Artificial Intelligence</strong></p> <p align="LEFT">Heb jij een passie voor IoT en zie je het als een uitdaging om die passie in te zetten in een onderzoek voor de agrarische sector? Onderzoek dan tijdens jouw afstudeeropdracht de mogelijkheden om met behulp van kunstmatige intelligentie geautomatiseerde inspecties van gewassen uit te voeren en adviezen te geven over bewatering, bemesting en het oogsten van deze gewassen.</p> <p align="LEFT">Het inspecteren van gewassen op akkers of in kassen is een tijdrovende klus. Daarnaast is de inspectie niet objectief, maar afhankelijk van de kennis en ervaring van de inspecteur. Hierdoor kunnen verkeerde beslissingen worden genomen met betrekking tot bemesting, bewatering en oogsten van gewassen, hetgeen hoge kosten met zich mee kan brengen.</p> <p align="LEFT">Jij gaat aan de slag met het vraagstuk hoe inspecties van gewassen geautomatiseerd uit te voeren zijn.</p> <p align="LEFT"><u>Onderzoek</u></p> <p align="LEFT">Tijdens jouw afstudeeropdracht onderzoek je de mogelijkheden om met behulp van kunstmatige intelligentie geautomatiseerde inspecties van gewassen uit te voeren en adviezen te geven over bewatering, bemesting en het oogsten van deze gewassen. Je implementeert een Proof of Concept (PoC) waarin je aantoont dat de oplossing die je bedacht hebt, werkt. Je toont aan dat deze in staat is de status van planten te bepalen en een advies te geven over bewatering, bemesting en oogst.</p> <p align="LEFT">Je gaat deze opdracht samen met een van onze engineers verder uitwerken. Dit met als doel om tot een definitieve afstudeeropdracht, compleet met deliverables, te komen die bij jou past.</p> ...
          Stage: Afstudeeropdracht richting continous delivery in Veenendaal      Cache   Translate Page      
<p align="LEFT"><strong>Smells like team spirit</strong></p> <p align="LEFT">Ben jij niet alleen bezig met het schrijven van code die er toe doet, maar ben jij je er ook van bewust hoe impact vaak gemaakt wordt in een goed en sfeervol team?</p> <p align="LEFT">Binnen een team is het belangrijk dat de sfeer onderling goed is. Hierdoor wordt commitment gecreëerd en wordt er beter samengewerkt. Dit komt de kwaliteit van het product ten goede. Bij outsourced teams is het soms lastig te bepalen hoe goed het team in zijn vel zit. Door de afstand en cultuurverschillen is het mogelijk dat er openheid ontbreekt. Dit zorgt ervoor dat er laat of niet bijgestuurd kan worden en er onderlinge conflicten kunnen ontstaan binnen het team.</p> <p align="LEFT">Als software ontwikkelbedrijf zijn wij constant op zoek naar manieren om processen te automatiseren. Ook hier vragen wij ons af of we op basis van gegevens uit ontwikkeltools kunnen bepalen wat de stemming binnen het team is. Wat is de ontwikkeling van de stemming binnen een team en wat is het effect hiervan op de resultaten? Denk bijvoorbeeld aan analyse op het commitgedrag, de teksten in commits, pull requests, discussies in userstories, etc.</p> <p align="LEFT"><u>Opdracht</u></p> <p align="LEFT">Binnen onze organisatie is veel data beschikbaar van projecten die in het verleden zijn uitgevoerd. Gebruik deze data om onderzoek te doen naar verschillende soorten input en tools die we kunnen gebruiken om de stemming in een team zo goed mogelijk te bepalen. Kijk daarnaast of er een voorspelling gedaan kan worden van het effect op het teamresultaat zodra de stemming in het team om dreigt te slaan. Denk bijvoorbeeld aan Azure Cognitive Services, Text Analytics API en Machine Learning.</p> <p align="LEFT">Bouw een Proof of Concept waarin data uit verschillende ontwikkeltools op een generieke manier kan worden opgehaald. Aggregeer en analyseer deze data om een inschatting te doen van de stemming binnen het team. Gebruik data van historische projecten om het Proof of Concept te testen.</p> ...
          Stage: Afstudeeropdracht richting NET, C#, Big Data, Machine Learning in Veenendaal      Cache   Translate Page      
<p align="LEFT"><strong>Betere energievoorziening met Distributie Automatisering Light</strong></p> <p align="LEFT">Ben jij geïnteresseerd in energie en Big Data? En wil je een bijdrage leveren aan een betrouwbaarder energienet in Nederland?</p> <p align="LEFT">Via een grotendeels onzichtbaar netwerk van kabels en leidingen voorziet Enexis als netbeheerder 2,7 miljoen huishoudens en bedrijven in het noorden, oosten en zuiden van Nederland van elektriciteit en gas.</p> <p align="LEFT">Netbeheerder Enexis heeft begin 2015 besloten een project DALI (Distributie Automatisering Light) te starten met als uiteindelijke doel observering van alle middenspanning (MS) stations (35.000 stuks) in 2020. Bij dit DALI programma worden de MS stations voorzien van een intelligent kastje voor monitoring en sturing waarmee o.a. openbare verlichting kan worden aangestuurd, trafo kWmax kan worden uitgelezen en informatie van kortsluitverklikkers richting het Bedrijfsvoeringscentrum (BVC) gestuurd kan worden.</p> <p align="LEFT">Bovengenoemde ontwikkelingen veroorzaken een toename van sensordata. Met de juiste ICT-voorzieningen levert deze bron van informatie nieuwe mogelijkheden om vroegtijdig onderbrekingen te identificeren en aan de hand van verbeterde analyse specifieker onderhoud te initialiseren. Dit alles om enerzijds sneller te acteren bij opgetreden energieonderbrekingen en anderzijds energieonderbrekingen nog beter te voorkomen.</p> <p align="LEFT"><u>Opdracht</u></p> <p align="LEFT">Bij DALI gaat software gebruikt worden die mede door ons ontwikkeld is om de sensordata te collecteren. Alle data wordt momenteel opgeslagen in een ongestructureerde datastore. En dan kom jij in beeld!</p> <p align="LEFT">Je onderzoekt (in overleg met een Business Expert) op welke wijze de data efficiënt omgezet kan worden naar waardevolle informatie.</p> <p align="LEFT">Vervolgens bewijs je de resultaten van het onderzoek door een Proof of Concept op de Microsoft Azure stack te bouwen, die informatie extraheert uit de gecollecteerde sensordata. Visualiseer vervolgens de informatie in een dashboard.</p> ...
          (USA-VA-Herndon) Senior Product Manager: SaaS | Software Products      Cache   Translate Page      
Senior Product Manager: SaaS | Software Products Senior Product Manager: SaaS | Software Products - Skills Required - Product Management, Designing Product Management Programs, SAAS Product Management, Software Product Management, Cybersecurity, Insurance, Platform Data, Methodology Design, Scoping, Designing and Implementing Product Features If you're a Senior Product Manager with SaaS and/or Software product experience, please read on! We apply artificial intelligence to solve complex, real-world problems at scale. Our Human+AI operating system, blends capabilities ranging from data handling, analytics, and reporting to advanced algorithms, simulations, and machine learning, enabling decisions that are just-in-time, just-in-place, and just-in-context. If this type of environment sounds exciting, please read on! **Top Reasons to Work with Us** - Benefits start on day 1 - Free onsite gym - Unlimited snacks and drinks - Located 1 mile from Wiehle-Reston East Station on the Silver line **What You Will Be Doing** RESPONSIBILITIES: The Senior Product Manager is responsible for the development of our portal which will be the primary interface for all our products. The Product Manager will be responsible for ensuring the products, features, and data used across our platform meet the needs of our business owners across the organization as well as our partners. The Product Manager must be passionate about user experience and be able to accommodate several different product use cases inside of one primary user interface. This will include not only utilizing our products and services, but creating an experience whereby a customer can build and distribute new products and services via our operating system. Specifically, our Senior Product Manager will lead all Product functions for our operating system. This will include managing product development for the development, integration of our products and services and the development of a third-party developer network to allow third parties to create and distribute applications. It will also require maintaining and augmenting each product over time. Our ideal candidate will be an expert in product management with a track record of success within the industry. The Senior Product Manager will be a process expert regarding the life-cycle of Product Management and the art and science of developing products that are demanded by customers and make meaningful improvements to the risks and challenges borne by our customers. The Senior Product Manager will also have a strong track record for methodology design, scoping, designing and implementing product features and managing a thoughtful product road-map. The Senior Product Manager will be very comfortable working in a fast-moving, dynamic environment with a tremendous amount of autonomy and keen focus on customer success. - Build vision, strategy, and road-map for product(s) with an end goal of creating a scalable and sustainable business line and meeting strategic objectives - Leading product design and product development road-map (e.g. gathering customer requirements, prioritizing the development of new features, sourcing partners, defining core capabilities needed to successfully roll-out the product/service internationally, etc.) - Develop a process to manage the product road-map for the portfolio of products and provide thoughtful guidance to their integration into a customer's environment - Develop a process and methodology for intake of internal and external feedback regarding requested features - Articulate a scoring model for feature development to maximize the company's growth while artfully planning the feature development cycle on a quarterly basis - Develop the business strategy and the marketing vision of products and services within the Platforms & Data team - Lead the business case for product feature development, ensuring sales department commitment and contemplating feature profitability - Own the customer experience with a strong focus on customer retention and expanding product adoption - Identify key market opportunities and capabilities where the data offering can deliver value to customers and gain traction - Develop programs and interactions with customers and advisory boards to maintain strong working relationships with enterprise customers to proactively identify opportunities for improvement and value creation - Understand customer needs, competitive landscape and the latest technology trends, and use this information to inform our product roadmap and development cycles - Drive cross-functional initiatives & alignment with key partner organizations (e.g., product marketing and sales) to ensure strong go-to-market cadence across all data offerings - Ensure there is a constant feedback mechanism in place regarding data and our internal infrastructure “practice what we preach” - Represent the customer for the development team - Creating materials for sales, marketing and clients (e.g. pricing, presentations, etc.) - Support the sales and strategy teams for business development and bids **What You Need for this Position** QUALIFICATIONS: - Bachelor's Degree - Capable of spending significant time with external and internal customers to gain both a deep understanding of customer needs and internal capability as required with product design and development teams - 5+ years of relevant professional experience directly related to Product Management and Data - Experience working with customers and development programs from small to large enterprises - Experience with designing a Product Management program from ground level to functional deployment at a company - A process-based approach to Product Management methodology deployment in all areas while being rigorous and passionate about process and methods - Effective organizational skills with strong attention to detail - Exceptional communication skills - both written and verbal **What's In It for You** - Competitive Salary - Incentive Stock Options - Medical, Dental & Vision Coverage - 401(K) Plan - Flexible “Personal Time Off (PTO) Plan - 10+ Paid Holiday Days Per Year So, if you're a Senior Product Manager with SaaS and/or Software product experience, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Senior Product Manager: SaaS | Software Products* *VA-Herndon* *WT1-1492954*
          (USA-VA-Herndon) Scrum Master: Startup | Small Business | $100K - $135K Salary      Cache   Translate Page      
Scrum Master: Startup | Small Business | $100K - $135K Salary Scrum Master: Startup | Small Business | $100K - $135K Salary - Skills Required - Scrum Master, Project Management, CSM | PMP-ACP | CSP Certification, Agile Principles | Practices | Theory, Online Products, Software as a Service (SaaS) Products, Agile Approaches (Scrum | Lean | XP | Kanban), Atlassian JIRA, PSM I | PSM II | SAFe, DevOps Tools and CI/CD Processes If you're a Certified Scrum Master with SMB and/or Startup experience, please read on! We apply artificial intelligence to solve complex, real-world problems at scale. Our Human+AI operating system, blends capabilities ranging from data handling, analytics, and reporting to advanced algorithms, simulations, and machine learning, enabling decisions that are just-in-time, just-in-place, and just-in-context. If this type of environment sounds exciting, please read on! **Top Reasons to Work with Us** - Benefits start on day 1 - Free onsite gym - Unlimited snacks and drinks - Located 1 mile from Wiehle-Reston East Station on the Silver line **What You Will Be Doing** RESPONSIBILITIES: As Scrum Master / Agile Lead, you will manage Agile-focused engineering development of our proprietary technology. You'll be working closely with the Product Manager and Engineer Lead, facilitating the Sprint process from beginning to end, ensuring the accurate and timely completion of product requirements. In addition, you'll drive the team's education in best practices of Agile development through a variety of methods meant to provide positive impact across the team. You should have a strong Project Management and/or Scrum Master background and want to be an active member of the team, not just someone who enforces Agile methodology. A background in working with online products and Software as a Service (SaaS) is strongly preferred, and the ability to handle the challenges faced by startups is key. Along with the Product and Engineering leads, you'll be part of the 3-person leadership team who will bring cutting edge and sometimes complex products to market, then identifying ways to evolve these products over time. You'll work closely with your leadership colleagues to ensure road-maps are built using reality as the driving force, not smoke and mirrors or unjustified optimism. - Lead diverse software development teams to on-time fulfillment of applications and services in a hybrid public-private cloud/computing environment - Facilitate the development and delivery of our proprietary products between Product and Engineering teams that drive market leading features into our customers hands - Act as a liaison for your team throughout the organization - Provide day-to-day oversight of the product road-map to ensure that timelines are met, tasks are prioritized appropriately, there is clear definition of scope and the team isn't over-committing - Ensure teams have appropriate direction and priority at all times to remain on track to reach internal and customer commitments and deadlines - Work with Product and Engineering to ensure the right level of understanding and the right culture/process exists to take high level concepts and feature ideas and help break them down into stories and sub-tasks - Track and report status of all projects you own during our weekly updates. In identifying areas where any risk may exist, proactively build, communicate and lead recovery actions - Participate in “scrum-of-scrum” meetings and assist in ongoing coordination of corporate wide road-maps to ensure alignment between teams across the organization - Manage the delivery of software enhancements and fixes per release schedules using Agile development methodologies - Help to shape and improve the product road-map by consolidating and communicating feedback captured during the execution and retrospective analysis of sprints and strategy meetings - Work with Product and Sales team to understand customer engagement needs and ensure product readiness is in alignment with road-map commitments and the teams ability to deliver - Work with Software Development leadership and teams to ensure technical requirements for engagement can be met on schedule; if necessary, identify and plan for any customer specific engineering requirements with Product - Follow customer integration needs, update implementation plans and report customer satisfaction outcomes internally to cross-functional teams - Ensure that all product milestones are tracking to schedule and communicate effectively to Product and or cross-functional teams whenever obstacles to meeting customer expectations surface **What You Need for this Position** QUALIFICATIONS: - Bachelor's Degree - 2+ years of experience as a full-time Scrum Master - Scrum Master certification (CSM) PMP-ACP or CSP certification - Expertise in diligently applying Agile principles, practices, and theory (User Stories, Story Pointing, Velocity, Burndown Metrics, Agile Games, Tasking, Retrospectives, Cycle Time, Throughput, Work in Progress levels, Product Demos) - Experience working with online products and Software as a Service (SaaS) - Experience working with different Agile approaches (Scrum, Lean, XP, Kanban, etc.) - Hands on experience with Agile ALM tools like Atlassian Jira with knowledge of other tools such as Aha!, Wrike, or ProductPlan - Knowledge of servant leadership, facilitation, situational awareness, conflict resolution, continual improvement, empowerment, influence and transparency - Agile or Project Management certification (PSM I, PSM II, SAFe) - Extensive experience with Atlassian products - Understanding of DevOps tools and CI/CD processes - Experience and interest in large-scale distributed systems - Interest in with Apache Spark, Apache Pig, AWS Pipelines, Google Dataflow, or MapReduce **What's In It for You** - Competitive Salary ($100,000 - $135,000) - Incentive Stock Options - Medical, Dental & Vision Coverage - 401(K) Plan - Flexible “Personal Time Off (PTO) Plan - 10+ Paid Holiday Days Per Year So, if you're a Certified Scrum Master with SMB and/or Startup experience, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Scrum Master: Startup | Small Business | $100K - $135K Salary* *VA-Herndon* *WT1-1492800*
          (USA-VA-Herndon) Full Stack Scala Engineer: JavaScript | Responsive Web Apps      Cache   Translate Page      
Full Stack Scala Engineer: JavaScript | Responsive Web Apps Full Stack Scala Engineer: JavaScript | Responsive Web Apps - Skills Required - JavaScript, Scala, Responsive Web Apps, Math, Modeling, JVM, Python, SPARK, Angular, Liftweb If you're an experienced Full Stack Scala Engineer, please read on! We apply artificial intelligence to solve complex, real-world problems at scale. Our Human+AI operating system, blends capabilities ranging from data handling, analytics, and reporting to advanced algorithms, simulations, and machine learning, enabling decisions that are just-in-time, just-in-place, and just-in-context. If this type of environment sounds exciting, please read on! **Top Reasons to Work with Us** - Benefits start on day 1 - Free onsite gym - Unlimited snacks and drinks - Located 1 mile from Wiehle-Reston East Station on the Silver line **What You Will Be Doing** RESPONSIBILITIES: - Design and develop code, predominantly in Scala, making extensive use of current tools such as Liftweb and Scala.js. - Developing state-of-the-art analytics tools supporting diverse tasks ranging from ad hoc analysis to production-grade pipelines and workflows for customer applications - Contributing to key user interactions and interfaces for tools across our modular SaaS platform - Developing tools to improve the ease of use of algorithms and data science tools - Working collaboratively to ensure consistent and performant approaches for the entire user experience and analytic code developed inside the system - Interacting directly with client project team members and operational staff to support live customer deployments **What You Need for this Position** QUALIFICATIONS: - Bachelor's Degree - Expert knowledge of Scala - Experience on full-stack software development teams - Expert knowledge of Javascript, HTML and CSS - Experience with responsive web applications - Experience with tools including Scala.js, Grunt, Bower, Liftweb - Advanced mathematical modeling skills - Experience with Akka, Akka HTTP, and Spark **What's In It for You** - Competitive Salary - Incentive Stock Options - Medical, Dental & Vision Coverage - 401(K) Plan - Flexible “Personal Time Off (PTO) Plan - 10+ Paid Holiday Days Per Year So, if you're an experienced Full Stack Scala Engineer, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Full Stack Scala Engineer: JavaScript | Responsive Web Apps* *VA-Herndon* *WT1-1492870*
          (USA-NY-New York) Machine Learning Engineer - NLP - New York      Cache   Translate Page      
Machine Learning Engineer - NLP - New York Machine Learning Engineer - NLP - New York - Skills Required - Machine Learning, Natural Language Processing, Artificial Intelligence If you are a Machine Learning Engineer with experience, please read on! We are recently closing a funding round and looking to hire for multiple position at our company! We are selling SaaS application to go on computers and enterprise accounts. We are about 100 person company with a bright future. Due to growth and demand for our services, we are in need of hiring for a Machine Learning Engineer that possesses strong experience with NLP and A.I. If you are interested in joining a small, growing company that pushes the envelope in the telecommunication industry and definitely cares about providing a great working environment for its employees, then apply immediately. **Top Reasons to Work with Us** 1. Work With Industry Leaders 2. joining a family of passionate, committed, success driven individuals 3. Opportunities for growth **What You Need for this Position** At Least 3 Years of experience and knowledge of: - Machine Learning - Natural Language Processing - Artificial Intelligence **What's In It for You** - Strong Annual Base Salary (D.O.E.) - Vacation/PTO - Comprehensive Benefit Plan So, if you are a Machine Learning Engineer with experience, please apply today! Please apply to job with most updated resume or send resume to: wendy.warner@cybercoders.com Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Machine Learning Engineer - NLP - New York* *NY-New York* *WW2-1493015*
          (USA-CA-San Jose) Lead Machine Learning Software Engineer - up to 220k + bonus      Cache   Translate Page      
Lead Machine Learning Software Engineer - up to 220k + bonus Lead Machine Learning Software Engineer - up to 220k + bonus - Skills Required - Machine Learning, Deep Learning, Natural Language Processing, Computer Vision, Programming with languages such as Python/C/Java, Full lifecycle experience Location: Work remote initially, once established office will be either Redwood City OR San Jose. Will need to be in the office 2-3 days per week minimum. Salary: Up to 220k base plus bonus Skills: Machine learning, full lifecycle experience, programming with a variety of languages Work for an industry leader which is one of the largest consumer products brands around the globe! It's an exciting time for our brand as we continue to move forward with our digital/IoT strategy. If you are a Lead Machine Learning Software Engineer please read on........ **Top Reasons to Work with Us** - Work for an industry leading consumer products brand - Excellent benefits including 401k contribution, bonuses and much more..... - Excellent work/life balance and positive company culture **What You Will Be Doing** As the Lead Machine Learning Engineer you will be very hands on defining and delivering solutions which will bring delightful user experiences globally. Key responsibilities: - Work with a cross functional team which is developing products for consumers across the globe - Utilize machine learning, computer vision, NLP and speech recognition techniques to create innovative products - Be the SME for Machine Learning in our product group - Stay abreast of the latest machine learning techniques and technologies and advise the company on how they can be applied to our products - Architect and implement smart IoT products - Mentor more junior engineers - Participate in code reviews **What You Need for this Position** Required: 5+ years in software engineering Strong Machine learning skills Programming with languages such as Python, C and Java Ideally you will have experience with at least some of these specific areas: computer vision, speech recognition, natural language processing **What's In It for You** Market rates salaries (150-220k) plus bonus and full benefits package! So, if you are a Lead Software Engineer that specializes in Machine Learning, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Lead Machine Learning Software Engineer - up to 220k + bonus* *CA-San Jose* *SJ2-LeadML-SJ*
          (USA-CA-San Jose) Lead Software Engineer: Embedded      Cache   Translate Page      
Lead Software Engineer: Embedded (IoT) Lead Software Engineer: Embedded (IoT) - Skills Required - RTOS, Embedded Android, IOT, Android, Linux, BSP, Android AOSP, AOSP, Android NDK, NDK Location: Work remote initially, once established office will be either Redwood City OR San Jose. Will need to be in the office 2-3 days per week minimum. Salary: Up to 220k base plus bonus Skills: system level embedded programming in Linux or Android or other lower layer system Work for an industry leader which is one of the largest consumer products brands around the globe! It's an exciting time for our brand as we continue to move forward with our digital/IoT strategy. If you are a Lead Embedded Engineer please read on.......... **Top Reasons to Work with Us** - Work for an industry leading consumer products brand - Excellent benefits including 401k contribution, bonuses and much more..... - Excellent work/life balance and positive company culture **What You Will Be Doing** The Lead Software Engineer will play a key role in the development of embedded software platforms to drive innovation across the full range of products. The software platforms range from RTOS to Embedded Android. The platforms provide IoT features, middleware to control appliance, framework for running machine learning models, and user-interface framework to allow development of interfaces. -Architect, design and develop software platforms using Android, Linux or RTOS -Lead the development of various software components that include interfacing to appliance control, appliance middleware, application layer networking protocols, device drivers, computer vision, deep learning inference middleware, GUI middleware, etc. -Working with partners to develop Android, Linux, or RTOS BSP including board-bringup, hardware debugging, and optimizing low-level OS features -Be the expert in Linux and Android System Software and especially develop expertise in Android's wireless and networking architecture, security architecture, low-power, over-the-air upgrades, and development tools -Provide technical leadership **What You Need for this Position** Required Qualifications: 5+ years in embedded engineering Android, Linux or RTOS Preferred Qualifications: -Experience working with the Linux open-source community is highly desired -Experience working within the Android development environment including with Android AOSP and Android NDK -Excellent with debugging of complex software systems -Have a deep understanding of full product development life-cycle **What's In It for You** Market rates salaries (150-220k) plus bonus and full benefits package! So, if you are a Lead Embedded Software Engineer with at least 5 years of experience, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Lead Software Engineer: Embedded* *CA-San Jose* *SJ2-Lead-IoT-SJ*
          (USA-PA-Pittsburgh) Computer Vision / Deep Learning Engineer      Cache   Translate Page      
Computer Vision / Deep Learning Engineer Computer Vision / Deep Learning Engineer - Skills Required - Computer Vision, C++, Caffe, Tensorflow, Lidar, Geometry-Based Vision, Deep Learning, Multi-view stereo I am currently working with several companies in the area who are actively hiring in the field of Computer Vision and Deep Learning. AI, and specifically Computer Vision and Deep Learning are my niche market specialty and I only work with companies in this space. I am actively recruiting for multiple levels of seniority and responsibility, from experienced Individual Contributor roles, to Team Lead positions, to Principal Level Scientists and Engineers. I offer my candidates the unique proposition of representing them to multiple companies, rather than having to work with multiple different recruiters at an agency, or applying directly to many different companies without someone to manage the process with each of those opportunities. In one example, I am working with a candidate who is currently interviewing with 10 different clients of mine for similar roles across the country with companies applying Computer Vision and Deep Learning to various different applications from Robotics, Autonomous Vehicles, AR/VR/MR, Medical Imaging, Manufacturing Automation, Gaming, AI surveillance, AI Security, Facial ID, 3D Sensors and 3D Reconstruction software, Autonomous Drones, etc. I would love to work with you and introduce you to any of my clients you see as a great fit for your career! Please send me a resume and tell me a bit about yourself and I will reach out and offer some times to connect on the phone! **Top Reasons to Work with Us** Some of the current openings are for the following brief company overviews: Company 1 - company is founded by 3x Unicorn (multi-billion dollar companies) founders and are breaking into a new market with advanced technology, customers, and exciting applications including AI surveillance, robotics, AR/VR. Company 2 - Autonomous Drones! Actually, multiple different companies working on Autonomous Drones for different applications - including Air-to-Air Drone Security, Industrial Inspection, Consumer Drones, Wind Turbine and Structure Inspection. Company 3 - 3D Sensors and 3D Reconstruction Software - make 3D maps of interior spaces using our current products on the market. We work with builders, designers, Consumers and Business-to-Business solutions. Profitable company with strong leadership team currently in growth mode! Company 4 - Industrial/Manufacturing/Logistics automation using our 3D and Depth Sensors built in house and 3D Reconstruction software to automate processes for Fortune 500 clients. Solid funding and revenue approaching profitability in 2018! Company 5 - Hand Gesture Recognition technology for controlling AR/VR environments. We have a product on the market as of 2017 and are continuing to develop products for consumers and business applications that are used in the real and virtual world. We have recently brought on a renowned leader in Deep Learning and it's intersection with neuroscience and are doing groundbreaking R&D in this field! Company 6 - Full facial tracking and reconstruction for interactive AR/VR environments. Company 7 - massively scalable retail automation using Computer Vision and Deep Learning, currently partnered with one of the largest retailers in the world. Company 8 - Products in the market including 3D Sensors, and currently bringing 3D reconstruction capabilities to mobile devices everywhere. Recently closed on a $50M round of funding and expanding US operations. Company 9 - Mobile AI company using Computer Vision for sports tracking and real time analytics for players at all levels from beginner to professional athletes to track, practice and improve at their craft. Company 10 - Digitizing human actions to create a massive new dataset in manufacturing - augmenting the human/robot working relationship and giving manufacturers the necessary info to improve that relationship. We believe that AI and robotics will always need to work side by side with humans, and we are the only company providing a solution to this previously untapped dataset! Company 11 - 3D facial identification and authentication for security purposes. No more key-fobs and swipe cards, our clients use our sensors and software to identify and permit employees. **What You Will Be Doing** If you are interested in discussing any of these opportunities, I would love to speak with you! I am interested in learning about the work you are currently doing and what you would be interested in for your next step. If the above opportunities are not quite what you're looking for but would still like to discuss future opportunities and potential to work together, I would love to meet you! I provide a free service to my candidates and work diligently to help manage the stressful process of finding the right next step in your career. The companies that I work with are always evolving so I can keep you up to date on new opportunities I come across. Please apply to this job, or shoot me an email at richard.marion@cybercoders.com and let's arrange a time to talk on the phone. **What You Need for this Position** Generally, I am looking for Scientists/Engineers in the fields of Computer Vision, Deep Learning and Machine Learning. I find that a lot of my clients are looking for folks who have experience with 3D Reconstruction, SLAM / Visual Odometry, Object Detection/Recognition/Tracking, autonomy, Point Cloud Processing, Software and Algorithm development in C++ (and C++11 and C++14), GPU programming using CUDA or other GPGPU related stuff, Neural Network training, Sensor Fusion, Multi-view stereo, camera calibration or sensor calibration, Image Segmentation, Image Processing, Video Processing, and plenty more! - Computer Vision - C+- Python - Linux - UNIX **What's In It for You** A dedicated and experienced Computer Vision placement specialist! If you want to trust your job search in the hands of a professional who takes care and pride in their work, and will bring many relevant opportunities your way - I would love to work with you! So, if you are a Computer Vision Scientist or Engineer and are interested in having a conversation about the market and some of the companies I am working with, please apply or shoot me an email with resume today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Computer Vision / Deep Learning Engineer* *PA-Pittsburgh* *RM2-1492758*
          (USA-VA-Springfield) Computer Scientist - Machine Learning      Cache   Translate Page      
Computer Scientist - Machine Learning Computer Scientist - Machine Learning - Skills Required - Computer Science, Machine Learning, large scale data, linear algebra, calculus If you are a Computer Scientist with a background in machine learning, please read on! You must have an active DoD secret clearance to be considered. This is a full time W2 position and great medical benefits are offered. This is not a contract. **What You Will Be Doing** As a lead technologist you will work with other scientists, engineers and performers to develop analytic capabilities for Military operations. - Conduct systems and application evaluations for developmental software systems - Consult with others to determine computing goals and systems requirements - Work with computer engineers and scientists to solve computing problems in relationship to threat spectrum - Use theoretical principles to guide development of software and hardware solutions - Help in the design of new computer technology - Use analytic programs, machine learning and statistical methods to prepare data for use in modeling - Devise new algorithms to solve problems **What You Need for this Position** More Than 5 Years of experience and knowledge of Computer Science: - A Ph.D in Computer Science is highly desired - Computer Science - Machine Learning - k-nearest neighbors, random forests, ensemble methods, etc. - CS related math skills - linear algebra, calculus, statistics and discrete mathematics - Complex problem solving skills - Machine learning techniques and frameworks - Can analyze large scale data - Design, validate and characterize algorithms and systems **What's In It for You** - Competitive compensation with excellent benefits - Long term job opportunity with growth potential - Work with a great group of colleagues that will support your work So, if you are a Computer Scientist with a background in machine learning, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Computer Scientist - Machine Learning* *VA-Springfield* *RK-1492789*
          (USA-NC-Cary) Software Engineer - PHP, Java, JavaScript      Cache   Translate Page      
Software Engineer - PHP, Java, JavaScript Software Engineer - PHP, Java, JavaScript - Skills Required - PHP, Java, JavaScript, MySQL, MVC, Linux, Apache, GIT, Subversion If you are a Software Engineer with PHP and Java experience, please read on! Based in Cary, NC - we provide accurate contact information and sales leads related to the e-commerce industry. Our company is looking for the right candidate to join our in-house engineering team. So, if you are interested in joining our growing team, please apply today! **Top Reasons to Work with Us** 1. Amazing Reputation. 2. Work with and learn from the best in the business. 3. Opportunity for career and income growth. **What You Will Be Doing** - Improve our next generation web technology tracking systems using machine learning and AI - Design and develop algorithms and techniques for high-volume data analysis - Build upon some of our most advanced platforms to maximize performance. **What You Need for this Position** At Least 3 Years of experience and knowledge of: - PHP - Java - JavaScript - MySQL - MVC - Linux - Apache - GIT - Subversion **What's In It for You** - Vacation/PTO - Medical - Dental - Vision - Relocation - Bonus - 401k So, if you are a Java Engineer with experience, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Software Engineer - PHP, Java, JavaScript* *NC-Cary* *JW8-1492814*
          Microsoft Azure Machine Learning and Project Brainwave – Intel Chip Chat – Episode 610      Cache   Translate Page      
In this Intel Chip Chat audio podcast with Allyson Klein: In this interview from Microsoft Ignite, Dr. Ted Way, Senior Program Manager for Microsoft, stops by to talk about Microsoft Azure Machine Learning, an end-to-end, enterprise grade data science platform. Microsoft takes a holistic approach to machine learning and artificial intelligence, by developing and deploying [...]
          (USA-MA-Waltham) Senior Java Developer      Cache   Translate Page      
Senior Java Developer Senior Java Developer - Skills Required - Java, NoSQL, JMS, Hibernate, Spring, JUnit, Hadoop, Cassandra, Solr, Jenkins Do you like tough problems? If so, we have an opportunity that will allow you to handle millions of customer requests per day all while making sense of a ton of data. Our current need is for a Senior Java Developer who possesses strong communication skills to join our growing team located near Newton, MA. Ideal candidates will be motivated to move into a leadership role once they have established themselves as a key player within our organization and we offer a ton of room for growth both technically and professionally. **Top Reasons to Work with Us** - Competitive Base Salary (150 - 170K) - Competitive Bonus Structure - Flexible Work Hours - 401K Plan - Extremely Competitive PTO Policy - Extreme Growth Opportunities and a fast track to leadership **What You Will Be Doing** -Be a major player in a company that's a pioneer in semantic technology -Work with cool technologies like Hadoop, Solr and Cassandra -Work with enormous data sets. Our database has over 10 billion records extracted from the Web -Learn data mining and machine learning techniques, such as Bayesian classifiers -Want to learn about what data science means in the real world? -Solve interesting and challenging problems alongside a great team of engineers -Develop new skills as you push your knowledge - and our technology - to new levels -Work for a profitable, growing company that works with an impressive Fortune 500 client list -Work on helping build/maintain/clean our platform of businessperson and company data, processing millions of records per day and billions of records overall. **What You Need for this Position** -Must have a strong knowledge of Java -Preferred experience with: Java EE, JMS, Hibernate, Spring, Junit -Eager to learn new technologies such as Hadoop, Cassandra, Solr, Jenkins, etc. -Interested in technologies/techniques like data mining, machine learning, clustering/tag clouds, etc. -Experience with NoSQL data stores is a plus -Experience with big data and data analysis or data science is a plus -Minimum 5-8 years experience in software development -A mindset that research should lead to actionable results -Excited to tell us why you want to work here and what kinds of challenges you're looking for -Can talk intelligently and passionately about the interesting challenges your projects presented -A sense of humor and perspective So, if you are a Lead Java Developer with experience, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Senior Java Developer* *MA-Waltham* *JW1-1492923*
          (USA-MO-St. Louis) Full Stack Developer      Cache   Translate Page      
Full Stack Developer Full Stack Developer - Skills Required - Python, React/Redux, Express/Electron, JavaScript, Semantic UI, Bootstrap, Data Analysis, PostgreSQL, Django, Flask If you are a Full Stack Javascript Developer with Python experience, you'll want to read this... We are a rapidly growing software startup headquartered in downtown Saint Louis. Using machine learning and speech-to-text AI technologies, we have created a product that transcribes and analyzes sales and customer service calls in real time to deliver live recommendations to representatives while they navigate their calls. We are looking for a Full Stack Engineer to join our exciting and rapidly growing team in Downtown St. Louis. **What You Will Be Doing** You will be creating, evolving, and maintaining our infrastructure which includes our desktop and web applications, cloud processing module, and transcription engine. **What You Need for this Position** - Full Stack Development - Javascript (React/Redux & Express/Electron) - Python (Flask, Tornado, Django, GRPC) - CSS (Semantic UI/Bootstrap) - PostgreSQL - Data Analysis **What's In It for You** - Competitive compensation with equity - Generous benefits package with the works - Work for a stable, rapidly growing company - Step into a high impact role - Relocation assistance available if needed - Surround yourself with top industry talent - Awesome office location in Downtown STL Incubator So, if you are a Full Stack Developer with experience, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Full Stack Developer* *MO-St. Louis* *GR1-1492920*
          (USA-WA-Bellevue) Machine Learning Scientist - NLP, Recommender/Ranking Systems      Cache   Translate Page      
Machine Learning Scientist - NLP, Recommender/Ranking Systems Machine Learning Scientist - NLP, Recommender/Ranking Systems - Skills Required - Machine Learning, NLP, Recommender Systems, Python, Deep Learning Theory, Hadoop, SPARK, Building Data Pipelines If you are a Machine Learning Scientist with experience, please read on! One of the largest and most well-known travel agencies is looking for a Machine Learning Scientist. We are an online travel agency that enables users to access a wide range of services. We books airline tickets, hotel reservations, car rentals, cruises, vacation packages, and various attractions and services via the world wide web and telephone travel agents. Our team helps power many of the features on our website. We design and build models that help our customers find what they want and where they want to go. As a member of our group, your contributions will affect millions of customers and will have a direct impact on our business results. You will have opportunities to collaborate with other talented data scientists and move the business forward using novel approaches and rich sources of data. If you want to resolve real-world problems using state-of-the-art machine learning and deep learning approaches, in a stimulating and data-rich environment, lets talk. **What You Will Be Doing** You will provide technical leadership and oversight, and mentor junior machine learning scientists You will understand business opportunities, identify key challenges, and deliver working solutions You will collaborate with business partners, program management, and engineering team partners You will communicate effectively with technical peers and senior leadership **What You Need for this Position** At Least 3 Years of experience and knowledge of: - PhD (MS considered) in computer science or equivalent quantitative fields with 3+ years of industry or academic experience - Expertise in NLP or recommender systems (strongly preferred) - Deep understanding of classic machine learning and deep learning theory, and extensive hands-on experience putting it into practice - Excellent command of Python and related machine learning/deep learning tools and frameworks - Strong algorithmic design skills - Experience working in a distributed, cloud-based computing environment (e.g., Hadoop or Spark) - Experience building data pipelines and working with live data (cleaning, visualization, and modeling) **What's In It for You** - Vacation/PTO - Medical - Dental - Vision - Bonus - 401k So, if you are a Machine Learning Scientist with experience, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Machine Learning Scientist - NLP, Recommender/Ranking Systems* *WA-Bellevue* *GK2-1493004*
          (USA-CA-Menlo Park) Fullstack Developer - Rails, React      Cache   Translate Page      
Fullstack Developer - Rails, React Fullstack Developer - Rails, React - Skills Required - Rails, REACT, Ruby On Rails, Python Title: Full Stack Developer Location: Menlo Park, CA Compensation: $140-160k- midlevel (4yrs); $160-190k- senior (6yrs+) PLUS EQUITY!! We're a fun, series B funded start-up and we work to bridge the gap between retailers and customers! We just received a massive round of funding and we're looking to grow our engineering team with some amazing fullstack developers! If you're looking to get in with a great company to grow with, please read on! **Top Reasons to Work with Us** -Laid back environment -Amazing company culture -Great benefits- full medical/dental/vision -Growing company -Unlimited PTO! -Fully stocked kitchen/catered lunches **What You Will Be Doing** 1.) Develop new features across our full stack and our iOS/Android apps 2.) Backup features w/corresponding tests 3.) Design, build and scale our web and mobile apps to be fast and reliable **What You Need for this Position** REQUIRED: 1.) 4+ years of professional, working experience with RAILS 2.) Experience with REST APIs 3.) Familiarity with the GIT workflow 4.) Experience with React, Python, Ruby Nice to haves (not required): 1.) Experience building native iOS or Android apps (personal projects count) 2.) DevOps skills- Amazon AWS 3.) Proficient in SQL, NoSQL 4.) Open-source projects on github 5.) Experience with R or Machine learning **What's In It for You** Aside from a competitive base salary, you will be rewarded with: 1.) GENEROUS equity in a FAST growing, early stage start-up 2.) Unlimited PTO 3.) Full Benefits- Med/Dental/Vision 4.) Fully stocked kitchen 5.) Catered lunches 6.) 10 Paid Holidays So, if you are a Fullstack Developer with experience, please apply today! Applicants must be authorized to work in the U.S. **CyberCoders, Inc is proud to be an Equal Opportunity Employer** All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, disability, protected veteran status, or any other characteristic protected by law. **Your Right to Work** – In compliance with federal law, all persons hired will be required to verify identity and eligibility to work in the United States and to complete the required employment eligibility verification document form upon hire. *Fullstack Developer - Rails, React* *CA-Menlo Park* *AH9-1492941*
          Dynamic Rule Based Decision Trees in Neo4j Part 4      Cache   Translate Page      

Dynamic Rule Based Decision Trees in Neo4j   Part 4

So far I’ve only showed you how to traverse a decision tree in Neo4j. The assumption being that you would either create the rules yourself from expert knowledge or via an external algorithm. Today we’re going to add an algorithm to build a decision tree (well a decision stream ) right into Neo4j. We will simply pass in the training data and let it build the tree for us. If you are reading this part without reading partsone,two, andthree, you should because this builds on what we learned along the way.

A decision tree is built with nodes that look at a value and go left if that value is less than or equal to some threshold, or go right if the value is greater. The nodes can only go left or right and can only go down one level at a time. Decision trees are a great starting point for machine learning models, but they suffer from a few problems: overfitting, instability and inaccuracy. These problems are overcome by combining a few hundred to several thousand decision trees together into a Random Forest . A random forest decreases the variance of the results without increasing the bias, which makes for a better model, but we have a very hard time looking at a random forest and understanding what it is really doing.

A decision stream allows nodes to follow a path based on multiple options and may go down more than 1 level. You can read the paper explaining what it is all about, but for our purposes, we are interested in knowing that a single decision stream can be as effective as a random forest, but a whole lot easier to understand. The authors of the paper were also gracious enough to code their algorithm for us to try out and that’s what we’ll do.

We are going to build a stored procedure that takes training data, answer data, and a significance threshold (which determines when to merge or split our nodes) and uses the resulting model to build a tree in Neo4j. Our training data is just a CSV file where the first row has a header and the following rows have numbers. If we had string data like “colors” where the options were “red, blue, yellow, etc” we would have to convert these to a number mapping 1 for red, 2 for blue, etc. For this project we are going to be reusing data from an old Kaggle competition that looked at the likeliness of someone defaulting on their loans.

RevolvingUtilizationOfUnsecuredLines,Age,Late30to59Days,DebtRatio,MonthlyIncome,OpenCreditLinesAndLoans,Late90Days,RealEstateLoans,Late60to89Days,Dependents
0.7661,45,2,0.803,9120,13,0,6,0,2
0.9572,40,0,0.1219,2600,4,0,0,0,1
0.6582,38,1,0.08511,3042,2,1,0,0,0

Our answer data is extremely simple, it’s just a single column of 1s and 0s for defaulted, and did not default:

Instead of diving into the stored procedure, I’m going to show you how to use it first. Follow the README, build the procedure and add it to Neo4j. We call it by giving it a few parameters. The name of the tree, the file where the training data lives, the file where the answers live and a threshold for merging and splitting rule nodes. In our case giving it a 0.02 which seemed like a good general value according to the paper:

CALL com.maxdemarzi.decision_tree.create('credit',
'/Users/maxdemarzi/Documents/Projects/branches/training.csv',
'/Users/maxdemarzi/Documents/Projects/branches/answers.csv', 0.02)

It takes 2-3 minutes to train this dataset of about 100k records and once it’s done we can see the results:


Dynamic Rule Based Decision Trees in Neo4j   Part 4

The tree node is in blue, the rules are green, our parameters are purple and our answers are in red. Notice that two of the parameter nodes “Monthly Income” and “Debt Ratio” are not connected to any rules. This tells us that these two values are not helpful in predicting the outcome, which kinda makes sense since these two parameters are used to qualify someone for a loan before they even get one. The first Rule node along the tree is the number of times someone is “Late 60 to 89 Days” paying their bills. Four different relationships emanate from there. Notice at the end when the Rule nodes connect to the Answer nodes they do so for both “IS_TRUE” and “IS_FALSE” relationships. I’ll explain this in a moment. First let’s try traversing the decision tree by passing in some values. This is the same procedure fromPart 3:

CALL com.maxdemarzi.decision_tree.traverse('credit',
{RevolvingUtilizationOfUnsecuredLines:'0.9572', Age:'40', Late30to59Days:'20',
DebtRatio:'0.1219', MonthlyIncome:'2600',OpenCreditLinesAndLoans:'4', Late90Days:'0',
RealEstateLoans:'0', Late60to89Days:'0', Dependents:'1'});
Dynamic Rule Based Decision Trees in Neo4j   Part 4

We get a path that ends in a Rule node checking “Age” connecting by the IS_TRUE relationship to both Answer nodes. The weights of those relationships however are different. We can see that if the user 66% likely to NOT default, vs 33% likely to default. So not only do we get a classifier, but we also get a confidence score.

If we omit the “Age” parameter in our query:

CALL com.maxdemarzi.decision_tree.traverse('credit',
{RevolvingUtilizationOfUnsecuredLines:'0.9572', Late30to59Days:'20',
DebtRatio:'0.1219', MonthlyIncome:'2600',OpenCreditLinesAndLoans:'4', Late90Days:'0',
RealEstateLoans:'0', Late60to89Days:'0', Dependents:'1'});
Dynamic Rule Based Decision Trees in Neo4j   Part 4

We get a partial path ending in the “Age” parameter as a way of asking for it, so we can ask the user for their age, re run the procedure and get a final answer.

The nice thing about this is that we can see and understand how the answer was derived. We can decide to alter the tree in any way, create many of them each from different training data, introduce parameters not in the original training set, whatever we want dynamically and still get results in real time.

I’m not going to explain the stored procedure line by line, but I do want to highlight a few things. You can see the whole thing at this repository . First thing is our stored procedure signature. Notice we are writing data to Neo4j so we need to use the Mode.Write option:

@Procedure(name = "com.maxdemarzi.decision_tree.create", mode = Mode.WRITE)
@Description("CALL com.maxdemarzi.decision_tree.create(tree, data, answers, threshold) - create tree")
public Stream<StringResult> create(@Name("tree") String tree, @Name("data") String data,
@Name("answers") String answers, @Name("threshold") Double threshold ) {

For all the different answers we are going to first create “Answer” nodes. In our case we only have 2 possibilities so, we will create two nodes.

for (Double value : answerSet) {
Node answerNode = db.createNode(Labels.Answer);
answerNode.setProperty("id", value);
answerMap.put(value, answerNode);
}

We want to create “Parameter” nodes for all the columns headers in our training data. We will save these in a “nodes” map and connect them to our Rules later.

HashMap<String, Node> nodes = new HashMap<>();
String[] headers = trainingData.next();
for(int i = 0; i < headers.length; i++) {
Node parameter = db.findNode(Labels.Parameter, "name", headers[i]);
if (parameter == null) {
parameter = db.createNode(Labels.Parameter);
parameter.setProperty("name", headers[i]);
parameter.setProperty("type", "double");
parameter.setProperty("prompt", "What is " + headers[i] + "?");
}
nodes.put(headers[i], parameter);
}

We will combine our answer and training data into a double array, which we then use to create a DoubleMatrix.

double[][] array = new double[answerList.size()][1 + headers.length];
for (int r = 0; r < answerList.size(); r++) {
array[r][0] = answerList.get(r);
String[] columns = trainingData.next();
for (int c = 0; c < columns.length; c++) {
array[r][1 + c] = Double.parseDouble(columns[c]);
}
}
DoubleMatrix fullData = new DoubleMatrix(array);
fullData = fullData.transpose();

The Decision Stream code was implemented in Clojure , but I don’t know Clojure so instead of trying to translate it into Java, I decided to just call it from our stored procedure. So we import Clojure core, get an interface to the training method for the model and then invoke it:

/* Import clojure core. */
final IFn require = Clojure.var("clojure.core", "require");
require.invoke(Clojure.read("DecisionStream"));
/* Invoke Clojure trainDStream function. */
final IFn trainFunc = Clojure.var("DecisionStream", "trainDStream");
HashMap dStreamM = new HashMap<>((PersistentArrayMap) trainFunc.invoke(X, rowIndices, threshold));

The training model returns as a nested hashmap with 4 values, the parameter, a threshold and two nested hashmaps on the left and right. From this we build our tree, combining leaf nodes whenever possible.

Node treeNode = db.createNode(Labels.Tree);
treeNode.setProperty("id", tree);
deepLinkMap(db, answerMap, nodes, headers, treeNode, RelationshipTypes.HAS, dStreamM, true);

The deepLinkMap method is used recursively for each side of the rule node, until we reach a Left node. One thing that was a bit of a pain was merging multi-option rule nodes into a single rule node, since the training map result doesn’t do this for us. The “merged” Rule nodes have a “script” property that ends up looking kinda like this:

if (Late60to89Days > 11.0) { return "IS_TRUE";}
if (Late60to89Days <= 11.0 && Late60to89Days > 3.0) { return "OPTION_1";}
if (Late60to89Days <= 3.0 && Late60to89Days > 0.0) { return "OPTION_2";}
if (Late60to89Days <= 3.0 && Late60to89Days <= 0.0) { return "OPTION_3";}
return "NONE";

Theoretically the “NONE” relationship type should never be returned, but the script needed a way to guarantee it ended and I didn’t want to mess with nested if statements.

Unmerged nodes have just two options “IS_TRUE” and “IS_FALSE” as well as a simple “expression” property that looks like the one below. The relationship type returned depends on the answer to the evaluation of that expression.

Late90Days > 0.0

As always the code is hosted on github , feel free to try it out, send me a pull request if you find any bugs or come up with enhancements. The one big caveat here is that I’m not a data scientists nor did I stay at a Holiday Inn Express last night , so please consult a professional before using.


          (USA-CA-San Diego) Systems Integration Engineer      Cache   Translate Page      
**Job Description** The Customer Technical Solutions team at Advanced GEOINT Systems (AGS) provides on-site installation, customization, support and training for the various GXP Platform, desktop and mobile products in environments ranging from traditional desktop to cloud based implementations. You will be required to conduct software demonstrations at trade shows, conferences and customer sites.You will also be responsible for providing incident management, problem management, and technical support for the customer environments while working closely with internal and external customers, internal support personnel, product testing, and engineering. As a member of the Customer Technical Solutions team, you will: + Work with members of our software support and test teams, product managers, and customers to integrate, configure, and apply intelligence processing, exploitation, and dissemination (PED) products. + Develop strong relationships with key customer technical personnel to ensure customer satisfaction and grow our business. + Become an expert in configuring and applying our products and provide customer feedback to product managers and the development teams. + Assist with training material development. **About GXP:** The Geospatial eXploitation Products (GXP) business provides licensed software capabilities and geospatial technology R&D. GXP s ability to draw on internal data production and technology expertise has allowed it to deliver superior products to the user community. GXP often finds ways to improve software implementation through user conferences and regional workshops, where important feedback and insight is gathered from customers. GXP commercial software, GXP Xplorer, GXP WebView, SOCET GXP, and SOCET SET provide customers with comprehensive image and video analysis, data management and geospatial production capabilities, andstate-of-the art sensor-data processing and analytics technologies, providing compact and scalable solutions to challenging problems in areas of video analytics, sensor data processing, computer vision, and machine learning. These products serve government and civil customers needs for photogrammetry, mapping, GIS, image exploitation, precision targeting, GEOINT (geospatial intelligence), MOVINT (movement intelligence), 3-D visualization, simulation and mission planning. **Culture characteristics:** + Treating customers, fellow employees, stakeholders, and partners with respect and dignity at all times. + Shows sincere commitment and adherence to all Corporate Compliance issues. + Humble attitude about knowledge limitations, know when to ask for help. + Ability to knowledge-share among team members, across partnered functions and the customer so that cases can continue with the utmost drive to a solution. + Have your peers' backs so that customers experience the same level of integrity regardless of whom they are working with. + Learn from one another. Take the time to learn a new product. + Demonstrate leadership in your actions. + Participating in maintaining an organized, clean, and safe work environment. + Self-discipline, the ability to prioritize tasks, and a detail-oriented working style. + Planning and managing projects without supervision. + Ability to learn from experience and informal instruction. + Adopt a strategy of continuous improvement. + In addition to having remarkable core IT skills, the Systems Integration Engineer must have customer focus in their DNA and do whatever it takes to achieve outstanding results for our customers. + Have a positive attitude, self-motivation, and a results-oriented approach to business. **General daily responsibilities include:** Performing technical planning, system integration, verification and validation, evaluates alternatives including cost, risk, supportability and analyses for total systems. Analysis is performed at all levels of total system product to include concept, design, fabrication, test, installation, operation, maintenance and disposal. Ensuring the logical and systematic conversion of product requirements into total systems solutions. Translating customer requirements into hardware/software specifications.Responsibilities will include installation and testing our solutions within the customer environment, to include: back-end services, front-end services, web applications, mobile applications, and improving the overall quality of our products. You will be required to analyze, test and verify the implementation of the functional and non-functional requirements of a cloud-based distributed system. + Submit defect reports and enhancement requests via JIRA. + Participate in Product Test activities. + Collaboratively work with Customer Technical Solutions team to resolve or diagnose customer issues. + Provide problem management by taking ownership of repeating incidents, researching root cause and resolving problem. + Understanding, applying, and communicating the technology concepts and implementations of our products. + Participating as a customer advocate in the engineering development process. + Contributing to the development of customer support material such as technical manuals, application briefs, and product data sheets as needed. **Typical Education & Experience** Typically a Bachelor's Degree and 4 years work experience or equivalent experience **Required Skills and Education** **Required Skills and Education** Bachelor's degree in computer science, engineering or another related fieldor at least four years of additional experience in lieu of a degree. Strong analytical, problem solving, and debugging skills. + Excellent communication skills, both written and verbal. + Excellent attention to detail, design, and user experience awareness. + Ability to work well in a very dynamic, fast moving environment with high expectations. + Flexibility in adjusting to variable workload and job duties. + Ability to work independently and with little supervision. + Proficiency with virtual servers (HyperV and VMware). + Experience installing, running, and troubleshooting software in Windows. + Exceptional organizational and prioritization skills. + Strong problem solving and analytical skills to diagnose issues at system level, including Windows/Linux operating system, SQL Server, web services and application integration. + Software diagnosis and troubleshooting. + A valid passport and the ability to travel domestically or internationally. + U.S. CITIZENSHIP REQUIRED. Candidates selected for some positions will be subjected to a government security investigation and will need to meet eligibility requirements for access to classified information. **Preferred Skills and Education** + SOCET GXP + GXP Xplorer + GEOINT + GIS + AWS + CITRIX XenServer / XenDesktop + VMWare VSphere, Horizon + NVIDIA GRID + Comfortable with presentations and speaking in front of large groups. + Multi-source processing, exploitation, and dissemination workflows. + Experience in providing software technical support. + Experience working with multiple OS types (Windows, Linux, OSX) and versions of SQL database configuration, optimization, and administration (MySQL, PostgreSQL). + Networking and networking protocols (TCP/IP, UDP, RDP). + Full-motion video (FMV) sensors, formats, codecs, and protocols. + Storage, retrieval, and streaming of video streams. + Programming in JavaScript, CSS, HTML, Python, and SQL. + Familiarity with OGC standards, such as WMS, WMTS, and WFS. **About BAE Systems Electronic Systems** BAE Systems is a premier global defense and security company with approximately 90,000 employees delivering a full range of products and services for air, land and naval forces, as well as advanced electronics, security, information technology solutions and customer support and services. The Electronic Systems (ES) sector spans the commercial and defense electronics markets with a broad portfolio of mission-critical electronic systems, including flight and engine controls; electronic warfare and night vision systems; surveillance and reconnaissance sensors; secure networked communications equipment; geospatial imagery intelligence products and systems; mission management; and power-and energy-management systems. Headquartered in Nashua, New Hampshire, ES employs approximately 13,000 people globally, with engineering and manufacturing functions primarily in the United States, United Kingdom, and Israel. Equal Opportunity Employer/Females/Minorities/Veterans/Disabled/Sexual Orientation/Gender Identity/Gender Expression **Systems Integration Engineer** **BAE1US21211** EEO Career Site Equal Opportunity Employer. Minorities . females . veterans . individuals with disabilities . sexual orientation . gender identity . gender expression
          Red Points, genuina ambición global frente a las falsificaciones      Cache   Translate Page      
«Queremos ser la empresa referente en el mundo en la lucha contra el fraude online». Esa es la ambiciosa declaración de intenciones que lanza la navarra Laura Urquizu, CEO de Red Points, una pujante startup española con sede en Barcelona y vocación global que ofrece soluciones tecnológicas contra la piratería y las falsificaciones en el inmenso bazar de Internet. La compañía fue creada en 2011 por los emprendedores Josep Coll y David Casellas, pero apenas era un embrión cuando Urquizu se sumó al proyecto. Con experiencia en el mundo financiero y en el ecosistema del emprendimiento digital (fue responsable del fondo de capital semilla de Caja Navarra, entidad de la que era miembro del comité de dirección), Urquizu quedó deslumbrada con las posiblidades del proyecto Red Points. La sintonía con sus fundadores hizo el resto. Urquizu decidió dar un giro a su carrera y aceptar la propuesta de convertirse en la CEO de una startup que parecía estar en el momento oportuno, con la tecnología necesaria y en el sitio adecuado: un mercado potencial gigantesco. Las expectativas no resultaron falsas. Red Point es uno de los singulares casos de startup capaz de pasar en solo dos años de uno a diez millones de dólares en ventas. En el foco de los grandes fondos globales, ha cerrado este año financiación por más de veinte millones de dólares para apuntalar sus capacidades tecnológicas y romper nuevos techos en su rápida ascensión comercial, como la apertura de una oficina en Nueva York. «Cuando los conocí aún era un proyecto pequeño que luchaba por crecer en el entorno de la piratería online y las descargas ilegales. Sus clientes eran productoras de cine, revistas... Pensé que tenían una oportunidad gigante si sumaban a su campo de acción la protección de las compañías contra los fraudes en la distribución online. Con eso ampliamos el market potencial en unas cien veces», explica Urquizu. La oportunidad era visible y estaba bien identificada. Urquizu no tuvo dudas en que era el momento de perseguirla: «Dejé todo lo que estaba haciendo, absolutamente todo. Era s eptiembre de 2014 y lo primero que hice fue reestructurar la empresa y potenciar la plataforma tecnológica, que ya presta servicios a más de 400 clientes», explica. Es una de las escasas startup que ha conseguido pasar de un millón a diez millones de facturación en dos años La inversión en tecnología es clave para responder a unas amenazas cada vez más sofisticadas. «Yo suelo comparar la piratería con un incendio en un bosque. Lamentablemente se extiende con gran rapidez, y no puedes hacer nada con cubos de agua, del mismo modo que un problema en Internet ya no se puede abordar de modo manual, solo con tencología», remarca. «Nosotros estamos continuamente aportando inteligencia a nuestra tecnología con desarrollos de "machine learning". La propia tecnología aprende, de forma que cuando lanzamos nuestros buscadores son cada vez más inteligentes», precisa la CEO de Red Points. La firma tiene un buscador especializado para cada marketplace (Alibaba, Aliexpress, eBay...) que rastrea las posibles falsificaciones. Tras un proceso de validación, la herramienta puede, con el permiso de la plataforma, eliminar el enlace que ofrece la venta de material fraudulento. Un proceso automatizado similar a al que se establece con la piratería de contenidos. La firma ha logrado 20 millones este año de fondos que respaldaron a firmas como Spotify o Alibaba La CEO de Red Points reconoce que gestionar un crecimiento tan rápido no es sencillo. «No solo es el número de clientes o el desarrollo de tecnología, también es el crecimiento en personas, porque al final una empresa de tecnología no es nada sin sus personas. El equipo es lo más importante, y digerir un crecimiento en el que se duplican cada año las personas del equipo (serán 170 a finales de 2018 frente a la docena de 2014) es complejo. Hay que luchar mucho, integrar equipos y buscar el mejor talento para crecer más, pero reconociendo el que ya tienes dentro, que es el que te ha hecho llegar hasta donde has llegado», concluye. La visión global de la empresa se traduce en la convivencia de empleados de 18 nacionalidades. Una vocación que «nos llevó a vender desde el minuto uno a cualquier parte del mundo. Solo el 5% de nuestros clientes es nacional», subraya Urquizu. Y que se extiende en la búsqueda de finaciación. «Desde el comienzo buscamos el apoyo de fondos internacionales que nos aportan algo más que dinero», afirma. Entre ellos está Mangrove, el fondo luxemburgués que apostó por Skype y Wix. «Es un fondo que se involucra con el CEO, que ayuda mucho, aportando todo su networking», afirma Urquizu. Northzone, el fondo sueco que dio alas a Spotify, o Eight Roads Ventures, el brazo inversor de Fidelity que impulsó a Alibaba, también han respaldado un proyecto antipiratería tocado por una genuina ambición.
          OM in the News: Making Sense of Supply Chain 4.0      Cache   Translate Page      
McKinsey, Cap Gemini and the Boston Consulting Group all suggest Supply Chain 4.0, digital transformation, is about applying digital technologies– Artificial Intelligence (AI), Machine Learning (ML), the Internet of Things (IoT) and Blockchain– to operational processes and creating improvements.  If digital transformation is to “transform” SCM, then it must as efficiently as possible match supply […]
          Support system for ATLAS distributed computing operations      Cache   Translate Page      
The ATLAS distributed computing system has allowed the experiment to successfully meet the challenges of LHC Run 2. In order for distributed computing to operate smoothly and efficiently, several support teams are organized in the ATLAS experiment. The ADCoS is a dedicated group of shifters who follow and report failing jobs, failing data transfers between sites, degradation of ATLAS central computing services, and more. The DAST provides user support to resolve issues related to running distributed analysis on the Grid. The CRC maintains a global view of the day- to-day operations. In this paper, the status and operational experience of the support system for ATLAS distributed computing in LHC Run 2 are reported. This report also includes operation experience from the Grid site point of view, and an analysis of the errors that create the biggest waste of wallclock time. The report of operation experience will focus on some of the more time-consuming tasks for shifters, and on the introduction of new technologies, such as machine learning, to ease the work.
          Data Architect/Data Science      Cache   Translate Page      
CA-SAN JOSE, Role : Data Architect/Data Science Location : San Jose California Duration : 6+ Months Expert programming skills in Python, R Experience in writing code for various Machine learning algorithms for classification, clustering, forecasting, regression, Neural networks and Deep Learning Hands-on experience with modern enterprise data architectures and data toolsets (ex: data warehouse, data marts, dat
          C/C++ Software Engineer/Machine Learning - Mobica - Warszawa, mazowieckie      Cache   Translate Page      
I hereby agree for processing of personal data by Mobica Limited with headquarter in Crown House, Manchester Road, Wilmslow, UK, SK9 1BH whose representative is...
Od Mobica - Fri, 19 Oct 2018 09:05:38 GMT - Pokaż wszystkie Warszawa, mazowieckie oferty pracy
          Senior C/C++ Software Engineer/Machine Learning - Mobica - Warszawa, mazowieckie      Cache   Translate Page      
I hereby agree for processing of personal data by Mobica Limited with headquarter in Crown House, Manchester Road, Wilmslow, UK, SK9 1BH whose representative is...
Od Mobica - Fri, 14 Sep 2018 21:05:32 GMT - Pokaż wszystkie Warszawa, mazowieckie oferty pracy
          Senior Site Reliability Engineer - Sift Science - Seattle, WA      Cache   Translate Page      
The Sift Science Trust PlatformTM uses real-time machine learning to accurately predict which users businesses can trust, and which ones they can't....
From Sift Science - Sun, 21 Oct 2018 06:15:49 GMT - View all Seattle, WA jobs
          Senior Software Development Engineer - Distributed Computing Services (Hex) - Amazon.com - Seattle, WA      Cache   Translate Page      
Knowledge and experience with machine learning technologies. We enable Amazon’s internal developers to improve time-to-market by allowing them to simply launch...
From Amazon.com - Thu, 26 Jul 2018 19:20:25 GMT - View all Seattle, WA jobs
          Machine Learning Engineer - Stefanini - McLean, VA      Cache   Translate Page      
AWS, Spark, Scala, Python, Airflow, EMR, Redshift, Athena, Snowflake, ECS, DevOps Automation, Integration, Docker, Build and Deployment Tools Ability to provide...
From Indeed - Tue, 16 Oct 2018 20:59:09 GMT - View all McLean, VA jobs
          Product Manager, Data - Upserve - Providence, RI      Cache   Translate Page      
Some of the technologies in our stack include React / React Native, Kinesis, DynamoDB, Machine Learning, RedShift, RDS, Ruby, JavaScript, Docker, ECS, and more....
From Upserve - Wed, 10 Oct 2018 18:03:02 GMT - View all Providence, RI jobs
          Executive Director- Machine Learning & Big Data - JP Morgan Chase - Jersey City, NJ      Cache   Translate Page      
We would be partnering very closely with individual lines of business to build these solutions to run on either the internal and public cloud....
From JPMorgan Chase - Thu, 01 Nov 2018 11:32:47 GMT - View all Jersey City, NJ jobs
          Sr Data Scientist Engineer (HCE) - Honeywell - Atlanta, GA      Cache   Translate Page      
50 Machine Learning. Develop relationships with business team members by being proactive, displaying a thorough understanding of the business processes and by...
From Honeywell - Thu, 20 Sep 2018 02:59:11 GMT - View all Atlanta, GA jobs
          Chief Strategist for Solutions and Data Architecture - HP - Palo Alto, CA      Cache   Translate Page      
Leverage Machine Learning/AI and NLP components in a solution. University degree in Computer Science or Engineering with great business understanding....
From HP - Tue, 16 Oct 2018 11:40:45 GMT - View all Palo Alto, CA jobs
          SAS Customer Intelligence 360: A look inside the black box of machine learning [Part 2]      Cache   Translate Page      

In part one of this blog posting series, we introduced machine learning models as a multifaceted and evolving topic. The complexity that gives extraordinary predictive abilities also makes these models challenging to understand. They generally don’t provide a clear explanation, and brands experimenting with machine learning are questioning whether they [...]

The post SAS Customer Intelligence 360: A look inside the black box of machine learning [Part 2] appeared first on SAS Blogs.


          Automatisez vos taches répétitives et à faibles valeurs ajoutées avec le RPA      Cache   Translate Page      
Le Robotic Process Automation automatise des processus métiers simples comme copier-coller, consolider ou extraire des données. Les collaborateurs sont alors délestés de leurs taches rébarbatives et sans valeurs ajoutées. En associant au RPA des solutions d'Intelligence artificielle, les taches exécutées sont plus complexes. Le collaborateur est alors augmenté et l'entreprise gagne en productivité, en sécurité et en traçabilité des opérations. Si l'IA et le big data font le buzz, le RPA ou Robotic Process Automation moins médiatisé en France permet d'accéder à ces technologies et de proposer de meilleurs services. Contrairement au nom qui peut laisser à penser qu'un robot humanoïde se cache derrière ce concept de RPA, il n'en n'est rien. Le Robotic Process Automation consiste à automatiser un processus métier via un logiciel, capable de suivre un schéma logique représentant graphiquement ce processus. Ainsi, le RPA permet d'automatiser des liaisons entre applications en évitant les ressaisies d'un applicatif à un autre, de consolider des comptes ou de réaliser des actions RH. Grâce au RPA, Coca Cola, automatise l'accueil d'un nouvel embauché en lui attribuant un badge, une adresse email, une entrée dans le logiciel de paye et des droits d'accès à certains applicatifs. De son côté Heineken consolide tous ces tableaux financiers partout dans le monde. Ce concept permet aussi de consolider des données clients, de vérifier la conformité d'un document, de renouveler une carte de crédit, de répondre à des demandes clients ou de gérer un processus depuis l'émission de la commande au paiement de la facture. En ciblant tout type de processus - achats, relation client, RH, production, juridique, comptabilité, etc- le RPA s'invite dans tous les secteurs d'activité et dans toutes les fonctions de l'entreprise. En quête de qualité, de productivité, de réduction des coûts et de solutions capables de les délester de toutes leurs tâches chronophages et sans valeur ajoutée, les entreprises sont de plus en plus nombreuses à s'intéresser à cette technologie. Selon Gartner, le marché mondial du RPA devrait atteindre 1 milliard de dollars en 2020 et, selon le cabinet MarketsandMarkets il devrait s'élever à 2,4 milliards de dollars d'ici 2022. Le RPA réduit le temps de traitement des dossiers Parmi les early adopters, se trouvent les secteurs de la banque, de la finance et de l'assurance. Ainsi elles utilisent déjà cette technologie dans le cadre de consolidation de comptes, de gestion de sinistres, de prêts immobiliers, ou de lutte contre la fraude ou le blanchissement d'argent. Agrave; titre d'exemple, la Bank of Ireland utilise le RPA pour faciliter la relation avec ses clients en agence. De son côté, la banque norvégienne BN BANK confie à un outil RPA l'étude de demandes de prêts immobiliers. Résultat : le délai de traitement d'un dossier est passé de 15 jours à une journée, et 126 000 dossiers ont été traités en un an pour un coût 100 fois inférieur au coût historique. Outre ces gains, le RPA est aussi un moyen de garantir la sécurité et la fiabilité d'exécution des process. En effet, les opérations réalisées par l'outil RPA sont enregistrées et auditables, et permettent d'identifier des activités frauduleuses ou de blanchissement d'argent. Grace au RPA, un conseiller clientèle pourra passer plus de temps à comprendre les besoins de son client et moins à collecter des documents, à les vérifier et à saisir des informations dans l'outil de la banque. De plus, l'outil de RPA apporte un haut niveau de sécurité puisque les opérations réalisées par les robots sont sécurisées, non répudiables et auditables, autant de points critiques pour assurer la traçabilité des opérations et des décisions. Le RPA : bénéfique pour l'emploi Dans le cas d'une déclaration d'accident de voiture, par exemple, l'assureur doit extraire du dossier de l'assuré le contrat lié au véhicule, vérifier le type de contrat, et rédiger un mail à l'assuré. Des tâches parfaitement exécutables par un automate dès lors que le cas entre dans le process pré-établi. En revanche, en cas de conditions ou de questions spécifiques, le conseiller reprend la main et fournit rapidement aux clients des réponses contextualisées et personnalisées. Avec le RPA les collaborateurs exécutent des tâches plus gratifiantes, plus créatives apportant ainsi un avantage compétitif à l'entreprise. Depuis les années 2000 les entreprises ont, pour rationaliser leurs dépenses, massivement externalisé dans les pays à faible coût de main d'oeuvre, toutes leurs tâches répétitives et chronophages réalisées manuellement. Mais avec le RPA, l'intérêt financier de l'offshore disparait. Selon une étude Everest Group, là où une entreprise faisait un gain de 28% en externalisant le traitement d'une feuille maladie, la RPA lui permet de réduire ses coûts de 50%. La tendance est donc à la relocalisation de tout ce type de prestations. Les taches automatisées se complexifient grâce à l'IA Couplée à l'intelligence artificielle (IA), le RPA va encore plus loin. Grâce au Machine Learning, les tâches automatisées sont de plus en plus complexes. En ajoutant des outils d'IA capables de repérer des émotions dans un mail (énervement, mécontentement, satisfaction, etc) l'automate peut sélectionner une réponse adaptée à la situation. Un process d'automatisation d'accord de crédit est également plus performant s'il est couplé à des outils d'OCR (Optical Character Recognition) capables de repérer un faux. Enfin, des outils de reconnaissance vocale ou de reconnaissance d'images permettront au RPA de traiter des documents non structurés. Au regard de tous ces bénéfices, le RPA est promis à un bel avenir. Il le sera d'autant plus s'il est associé à de l'intelligence artificielle. Après la robotisation des usines, l'automatisation intelligente entre dans les services !
          Nutzung von Machine Learning verzehnfacht sich in den nächsten vier Jahren      Cache   Translate Page      
The unbelievable Machine Company: Berlin (ots) - Studie belegt: Digitaler Wertschöpfungsanteil durch KI und Machine Learning nimmt rasant zu - allein die deutschen Top-100-Unternehmen setzen 2022 über 100 Mrd. Euro damit um Die Hälfte der Unternehmen in Deutschland beschäftigt sich ...
          The one video everyone should watch if they want to know what Machine Learning is      Cache   Translate Page      

Morgan Linton Morgan Linton: As the co-founder of a company with Machine Learning at its core I spend a lot of time explaining Machine Learning (ML) to people. While I have tried to really hone my approach to explaining machine learning in two minutes – the reality is, you need more time to truly understand ML, even the fundamentals. […]

The post The one video everyone should watch if they want to know what Machine Learning is appeared first on iGoldRush Domain News and Resources.


          Architect - Microsoft - Redmond, WA      Cache   Translate Page      
Preferred experience with at least one of the Machine Learning related technologies (SAS, SPSS, RevR, Azure ML, MapR)....
From Microsoft - Thu, 23 Aug 2018 08:25:24 GMT - View all Redmond, WA jobs
          UR - Corporate Engineering - Precision Systems Engineer (Maplewood, MN) - 3M - Maplewood, MN      Cache   Translate Page      
Proactively collaborate with business partners to connect and extend process data management solutions with complimentary machine learning and analytics efforts...
From 3M - Wed, 05 Sep 2018 17:09:45 GMT - View all Maplewood, MN jobs
          Data Science Manager - Micron - Boise, ID      Cache   Translate Page      
Create server based visualization applications that use machine learning and predictive analytic to bring new insights and solution to the business....
From Micron - Wed, 05 Sep 2018 11:18:49 GMT - View all Boise, ID jobs
          Intern - Data Scientist (NAND) - Micron - Boise, ID      Cache   Translate Page      
Machine learning and other advanced analytical methods. To ensure our software meets Micron's internal standards....
From Micron - Wed, 29 Aug 2018 20:54:50 GMT - View all Boise, ID jobs
          Intern - Data Scientist (DRAM) - Micron - Boise, ID      Cache   Translate Page      
Machine learning and other advanced analytical methods. To ensure our software meets Micron's internal standards....
From Micron - Mon, 20 Aug 2018 20:48:37 GMT - View all Boise, ID jobs
          Director of Artificial Intelligence - Micron - Milpitas, CA      Cache   Translate Page      
Broad, versatile knowledge of artificial intelligence and machine learning landscape, combined with strong business consulting acumen, enabling the...
From Micron - Fri, 19 Oct 2018 17:30:05 GMT - View all Milpitas, CA jobs
          Junior Machine Learning Developer - Gartner - Québec City, QC      Cache   Translate Page      
What makes Gartner a GREAT fit for you? When you join Gartner, you’ll be part of a fast-growing team that helps the world become smarter and more connected. We...
From Gartner, Inc. - Thu, 02 Aug 2018 14:12:39 GMT - View all Québec City, QC jobs
          Junior Machine Learning Developer - Gartner - Québec City, QC      Cache   Translate Page      
What makes Gartner a GREAT fit for you? When you join Gartner, you’ll be part of a fast-growing team that helps the world become smarter and more connected. We...
From Gartner, Inc. - Thu, 02 Aug 2018 14:12:39 GMT - View all Québec City, QC jobs
          ScoreData Brings AI and Dynamic Machine Learning™ to Talkdesk Through...      Cache   Translate Page      

Out-of-the-box AI-powered contact-center for a broad range of TalkDesk customers and partners

(PRWeb November 07, 2018)

Read the full story at https://www.prweb.com/releases/scoredata_brings_ai_and_dynamic_machine_learning_to_talkdesk_through_appconnect_partnership/prweb15895435.htm


          Graduate Development Program -- 4      Cache   Translate Page      
We are looking for a trainer to provide class room training to 20 university fresh graduates on the following subjects: Digital banking, AI Big data Machine learning Fintech Data Analytics Blockchain... (Budget: min $50000 USD, Jobs: Artificial Intelligence, Data Analytics, Data Mining, Digital Marketing, Machine Learning)
          Machine Learning Scientist - NLP, Recommender/Ranking Systems      Cache   Translate Page      
WA-Bellevue, If you are a Machine Learning Scientist with experience, please read on! One of the largest and most well-known travel agencies is looking for a Machine Learning Scientist. We are an online travel agency that enables users to access a wide range of services. We books airline tickets, hotel reservations, car rentals, cruises, vacation packages, and various attractions and services via the world wid
          Technical Architect - Data Solutions - CDW - Milwaukee, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Milwaukee, WI jobs
          Technical Architect - Data Solutions - CDW - Madison, WI      Cache   Translate Page      
Experience with Microsoft SQL, MySQL, Oracle, and other database technologies. Predictive analytics, Machine Learning and/or AI skills will be a plus....
From CDW - Sat, 03 Nov 2018 06:11:29 GMT - View all Madison, WI jobs
          Director, Marketing Automation for Relationship Marketing - Microsoft - Redmond, WA      Cache   Translate Page      
In addition, this role will lead the identification and leveraging of the latest marketing capabilities like machine learning and AI to propel Relationship...
From Microsoft - Wed, 31 Oct 2018 07:57:36 GMT - View all Redmond, WA jobs
          Business Analysis Manager - Business Intelligence - T-Mobile - Bellevue, WA      Cache   Translate Page      
Entrepreneurial spirit and interest in advance analytics, big data, machine learning, and AI. Do you enjoy using data to influence technology, operations and...
From T-Mobile - Wed, 10 Oct 2018 03:14:47 GMT - View all Bellevue, WA jobs
          RadarServices eröffnet neuen Forschungs- und Entwicklungsstandort in Wien      Cache   Translate Page      
Die Unternehmensbereiche Engineering und Research arbeiten ab sofort vom neuen Standort „Radar R&D Hub Vienna“ aus. Er befindet sich im 7. Wiener Gemeindebezirk, in unmittelbarer Nachbarschaft zum Headquarter von RadarServices im Herzen der Stadt. Im Mittelpunkt der Tätigkeiten stehen die Forschung im Bereich Machine Learning und die Weiterentwicklung der eigenen Technologie für das kontinuierliche Cyber […]
          Upaya Grab dan Go-Jek Atasi Fraud, dari GPS Palsu hingga Order Fiktif      Cache   Translate Page      

Liputan6.com, Jakarta - Permasalahan order fiktif hingga fake GPS yang dilakukan sejumlah mitra pengemudi layanan transportasi online harus diakui masih terjadi.

Kendati demikian, penyedia layanan seperti Grab dan Go-Jek bukannya tanpa usaha untuk mengatasi masalah tersebut.

Baik Grab dan Go-Jek mengaku sudah menerapkan sejumlah fitur dan program agar aksi curang tersebut dapat diatasi.

Saat dihubungi Tekno Liputan6.com, Rabu (7/11/2018), Grab mengatakan pihaknya sudah memanfaatkan teknologi machine learning (pembelajaran mesin) dan artificial intelligence (kecerdasan buatan/AI) untuk mendeteksi aplikasi fake GPS dan order fiktif.

"Ketika order demikian terdeteksi, akun yang menggunakan aplikasi (fake GPS) tersebut akan terblokir," tutur Head of Public Affairs Grab Indonesia Tri Sukma Anreianno.

Tidak hanya itu, Grab juga membuat program 'Grab Lawan Opik!' yang bekerja sama dengan Kepolisian.

Dalam program ini, Grab telah berhasil menangkap sindikat dan mitra pengemudi yang telah terbukti melakukan kecurangan di beberapa kota, seperti Jakarta, Makassar, Semarang, Surabaya, dan Medan.

Menurut Tri, gabungan program dan fitur ini terbukti telah mengurangi tindak kecurangan hingga 80 persen.

"Riset kami juga menunjukkan bahwa platform Grab dua kali lebih tangguh dalam menghadapi tindak kecurangan dibandingkan kompetitor lain di Asia Tenggara," tuturnya menjelaskan.

Tidak hanya itu, Grab juga mengambil sikap tegas terhadap pemesanan fiktif atau yang dikenal sebagai opik dengan melakukan akun mitra pengemudi.

Program ini, menurut Tri, merupakan komitmen Grab untuk menyediakan platform transportasi teraman bagi mitra pengemudi dan penumpang.

"Manajemen Grab tidak akan melakukan peninjauan ulang terhadap keputusan hubungan kemitraan (suspend) pada mitra pengemudi yang terbukti melakukan tindak kecurangan (fraud)," tutur Tri menjelaskan.

Fitur Grab untuk Atasi Fraud

Pengemudi ojek online menunggu penumpang di tempat drop off yang disediakan di Balai Kota DKI Jakarta, Selasa (31/7). Pengemudi ojek online harus menempati area tersebut saat menunggu penumpang. (Liputan.com/Faizal Fanani)#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

Dalam penjelasannya, Tri juga menuturkan Grab sudah menerapkan fitur 'Anti-Tuyul'. Fitur yang diluncurkan pada Agustus 2018 ini memungkinkan Grab memblokir mitra pengemudi yang memiliki aplikasi fake GPS atau yang lebih dikenal 'Tuyul'.

Untuk mendapatkan kembali akses terhadap akunnya, mitra pengemudi harus menghapus seluruh apliksi fake GPS yang dimilikinya. Sementara untuk layanan GrabCar, ada pula fitur 'driver selfie authentication'.

Melalui fitur ini, mitra pengemudi diwajibkan mengambil dan mengunggah swafoto dirinya sebelum memulai atau meneruskan perjalanan. Cara ini dilakukan untuk memastikan hanya pengemudi terverifikasi yang memakai akun tersebut.

Perusahaan yang berbasis di Singapura itu juga mengklaim menjadi satu-satunya perusahaan ride-hailing yang secara ketat menerapkan prinsip 'Know Your Driver Partner' (KYP) untuk mitra pengemudi GrabCar.

Prinsip ini dijalankan dengan memeriksa seluruh dokumen fisik yang dimiliki mitra, seperti KTP, SIM, STNK, SKCK hingga bertemu langsung.

Serupa dengan Grab, Go-Jek sebagai salah satu penyedia layanan transportasi berbasis aplikasi menolak penggunaan aplikasi fake GPS atau order fiktif.

Namun, Go-Jek menerapkan sistem yang berbeda. Hal itu dituturkan oleh VP Corporate Affairs Go-Jek Michael Reza Say.

"Go-Jek pada dasarnya selalu menjunjung prinsip keadilan dan kejujuran. Penggunaan aplikasi GPS palsu merugikan mitra sendiri dan juga mitra lainnya yang bekerja dengan jujur," tuturnya.

Oleh sebab itu, Go-Jek telah menjalankan kebijakan Hapus Tuyul atau menghapus aplikasi GPS palsu di Jakarta, Bandung, Pontianak, Tegal, Surabaya, Medan, dan Balikpapan sejak Maret 2018.

"Hal ini merupakan langkah awal kami untuk memastikan ruang bekerja yang bersih, jujur dan adil bagi para mitra. Hingga saat ini kami telah berhasil menurunkan angka pengguna GPS palsu lebih dari setengahnya," ujar Michael.

Langkah Go-Jek Lawan Fraud

Driver Grab Bike Malang#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

Dalam beberapa waktu ke depan, Go-Jek juga akan menerapkan sistem pengalokasian pengemudi yang diperbarui. Melalui cara ini, mitra pengemudi yang berlaku jujur akan mendapatkan lebih banyak pesanan.

Maksudnya, mitra pengemudi yang terdeteksi menggunakan aplikasi fake GPS atau aplikasi tambahan tidak resmi lainnya akan lebih sulit mendapatkan pesanan. Sementara mitra pengemudi yang tidak melakukannya akan lebih diuntungkan.

"Kami juga selalu mengimbau agar mitra driver menghentikan penggunaan GPS palsu karena dapat menganggu keamanan data dari akun mitra itu sendiri," tuturnya menjelaskan.

Dari sisi order fiktif, Michael menuturkan, sistem Go-Jek sudah lebih baik dalam mengidentifikasi dan menangani order fiktif. Menurut Michael, 90 persen pesanan fiktif telah berhasil dihentikan sebelum sampai ke aplikasi mitra pengemudi Go-Jek.

"Kami akan melihat pola anomali dari akun tersebut. Yang pasti, kami akan terus memperbarui sistem sehingga permasalahan ini dapat diselesaikan secara menyeluruh agar tidak mengganggu aktivitas para mitra," ucap Michael menutup pembicaraan.

Sekadar informasi, Grab dan Go-Jek harus diakui merupakan dua pemain utama di bisnis ride hailing di Indonesia. Grab sendiri kini sudah hadir di 137 kota di Indonesia, sedangkan Go-Jek telah berada di 70 lokasi.

(Dam/Isk)

Saksikan Video Pilihan Berikut Ini: 


          Isotonik Studios releases Factormini for Ableton by JJ Burred      Cache   Translate Page      
JJ Burred Factormini

Isotonik Studios has announced Factormini, an introductory version of the Factorsynth Max for Live device by JJ Burred. Shareing the same decomposition engine as its bigger brother, Factormini contains the essential features for sound deconstruction. Factormini is the introductory version of Factorsynth, a powerful new type of musical tool that brings machine learning to Ableton […]

The post Isotonik Studios releases Factormini for Ableton by JJ Burred appeared first on rekkerd.org.


          Product Operations Intern - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 17 Oct 2018 22:29:11 GMT - View all New York, NY jobs
          QA Engineer - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Sun, 07 Oct 2018 06:47:33 GMT - View all New York, NY jobs
          Data Scientist - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Thu, 04 Oct 2018 06:17:29 GMT - View all New York, NY jobs
          NLP Data Scientist - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Thu, 27 Sep 2018 14:36:48 GMT - View all New York, NY jobs
          Business Development Representative - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 26 Sep 2018 16:31:17 GMT - View all New York, NY jobs
          UI/UX Designer - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 26 Sep 2018 16:18:14 GMT - View all New York, NY jobs
          Customer Success Manager - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 26 Sep 2018 14:35:24 GMT - View all New York, NY jobs
          Software Architect - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 26 Sep 2018 14:35:24 GMT - View all New York, NY jobs
          UI Engineer - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Tue, 25 Sep 2018 06:27:49 GMT - View all New York, NY jobs
          Mid-Market Account Executive - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Mon, 24 Sep 2018 16:34:54 GMT - View all New York, NY jobs
          Full Stack Engineer - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 19 Sep 2018 06:16:43 GMT - View all New York, NY jobs
          Enterprise Account Executive - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Mon, 10 Sep 2018 14:31:19 GMT - View all New York, NY jobs
          MACHINE LEARNING INTERN FOR SPEECH RELATED APPLICATIONS - Huawei Canada - Montréal, QC      Cache   Translate Page      
We thank all applicants for their interest in career opportunities with Huawei. Machine Learning Intern for Speech related Applications....
From Huawei Canada - Wed, 17 Oct 2018 05:55:32 GMT - View all Montréal, QC jobs
          MACHINE LEARNING ENGINEER FOR SPEECH RELATED APPLICATIONS - Huawei Canada - Montréal, QC      Cache   Translate Page      
We thank all applicants for their interest in career opportunities with Huawei. Machine Learning Engineer for Speech related Applications (6 months contract)....
From Huawei Canada - Wed, 17 Oct 2018 05:55:32 GMT - View all Montréal, QC jobs
          Machine Learning Software Developer - Huawei Canada - Montréal, QC      Cache   Translate Page      
We thank all applicants for their interest in career opportunities with Huawei. ML Software developer....
From Huawei Canada - Fri, 05 Oct 2018 05:47:02 GMT - View all Montréal, QC jobs
          MACHINE LEARNING HARDWARE RESEARCHER OR DEVELOPER - Huawei Canada - Montréal, QC      Cache   Translate Page      
We thank all applicants for their interest in career opportunities with Huawei. Machine Learning Hardware Researcher or Developer....
From Huawei Canada - Fri, 05 Oct 2018 05:47:02 GMT - View all Montréal, QC jobs
          Data Elixir - Issue 207      Cache   Translate Page      

In the News

Harvard Converts Millions of Legal Documents into Open Data

Inspired by the Google Books Project, the new Caselaw Access Project from the Library Innovation Lab at Harvard puts the entire corpus of published U.S. case law online for anyone to access for free. The project involved scanning and digitizing 100,000 pages per day over two years. This is a big deal that will enable new analytical insights, research, and applications.

govtech.com

⚽ How data analysis helps football clubs make better signings

They said it could never be done. The game was too fluid, too chaotic. The players’ movements too difficult to track reliably. But, decades after sports like baseball first embraced statistics, football - known as soccer in the U.S. - is starting to play the data game.

ft.com

Sponsored Link

10 Guidelines for A/B Testing

Online experimentation, or A/B testing, is the gold standard for measuring the effectiveness of changes to a website. But while A/B testing can appear simple, there are many issues that can complicate an analysis. In this presentation, Emily Robinson, data scientist at DataCamp, will cover 10 best practices that will help you avoid common pitfalls.

gotowebinar.com

Tools and Techniques

Importance of Skepticism in Data Science

Great discussion about one of the most important aspects of analyzing data - being skeptical of the results. Includes lots of useful examples.

github.io

Scaling Machine Learning at Uber with Michelangelo

In 2015, machine learning was not widely used at Uber. Just three years later, Uber has advanced capabilities and infrastructure, and hundreds of production machine learning use-cases. This post describes the wide variety of ways that Uber uses machine learning and how they've managed to scale their systems so quickly and effectively.

uber.com

Why Jupyter is data scientists’ computational notebook of choice

Last week's article in Nature about Jupyter Notebooks sparked a great discussion in Hacker News. There's a lot here, including tips, debate, and links to further resources.

ycombinator.com

Grokking Deep Learning

Andrew Trask's new book, Grokking Deep Learning aims to be the easiest introduction possible to deep learning. Each section teaches how to build a neural component from scratch in NumPy. This repo contains the code examples for each lesson.

github.com

Bringing machine learning research to product commercialization

Rasmus Rothe, Founder at Merantix, explores the differences between academia and industry when applying deep learning to real-world problems. This article goes into detail about differences regarding workflow, expectations, performance, model design and data requirements.

medium.com

Find A Data Science Job Through Vettery

Vettery specializes in tech roles and is completely free for job seekers. Interested? Submit your profile, and if accepted onto the platform, you can receive interview requests directly from top companies growing their data science teams.

// sponsored

vettery.com

Resources

Data Science With R Workflow Cheatsheet

Nice map of R cheatsheets that's organized around common workflows. It's like a cheatsheet for cheatsheets.

business-science.io

Python Learning Resources

Rachel Thomas from fast.ai asked for Python learning recommendations on Twitter and the resulting thread was amazing. These two recommendations, in particular, stand out:

twitter.com

Career

My Weaknesses as a Data Scientist

Identifing your weaknesses is one of most important things you can do to become effective in your career. In this post, William Koehrsen explores his particular weaknesses as a data scientist and the steps he's taking to overcome them. The approach he models here may be uncomfortable for some but it's a super effective strategy.

towardsdatascience.com

Jobs & Careers

Hiring?

Post on Data Elixir's Job Board to reach a wide audience of data professionals.

dataelixir.com

Recent Listings:

More data science jobs >>

About

Data Elixir is curated and maintained by @lonriesberg. For additional finds from around the web, follow Data Elixir on Twitter, Facebook, or Google Plus.


This RSS feed is published on https://dataelixir.com/. You can also subscribe via email.


          Architect - Microsoft - Redmond, WA      Cache   Translate Page      
Preferred experience with at least one of the Machine Learning related technologies (SAS, SPSS, RevR, Azure ML, MapR)....
From Microsoft - Thu, 23 Aug 2018 08:25:24 GMT - View all Redmond, WA jobs
          AMD anunță noile plăci Radeon Instinct MI60 și MI50      Cache   Translate Page      
AMD anunță noile plăci Radeon Instinct MI60 și MI50

Primele plăci Radeon construite pe procesul de 7 nanometri au fost dezvăluite seara trecută de către AMD.

Acestea nu sunt pentru gaming, în cazul în care vă întrebați, ci fac parte din seria Instinct, gândită pentru accelerarea aplicațiilor din sectorul High Performance Computing și Machine Learning.

Citeste articolul pe site-ul ZONA


          The one video everyone should watch if they want to know what Machine Learning is      Cache   Translate Page      

Morgan Linton Morgan Linton: As the co-founder of a company with Machine Learning at its core I spend a lot of time explaining Machine Learning (ML) to people. While I have tried to really hone my approach to explaining machine learning in two minutes – the reality is, you need more time to truly understand ML, even the fundamentals. […]

The post The one video everyone should watch if they want to know what Machine Learning is appeared first on iGoldRush Domain News and Resources.


          GA & Data Mining Projects (in Python and C/C++)      Cache   Translate Page      
I am looking for someone to work on Algorithmic and Data Mining projects. Must understand how to program Genetic Algorithms (GA), the 8-queens problem and other data mining algorithms. Must be also good at Mathematics, Statistics and Computer Science to understand coding needs... (Budget: $10 - $30 USD, Jobs: C Programming, Computer Science, Data Mining, Machine Learning, Python)
          The Art of Artificial Intelligence      Cache   Translate Page      
This week on the Cultural Frontline we meet the artists working with AI to reimagine the worlds of visual art, music and movies. Scientist and AI aficionado Janelle Shane sorts AI science fiction from AI science fact and explains how this new technology is being harnessed by artists across the world. As an AI-generated piece of art goes under the auction hammer at Christie’s in New York for the first time, we meet two pioneering artists creating surreal visual art in collaboration with machines, Mario Klingemann and Harshit Agrawal. Is this the start of a new era in the art world? We discover what happened when Iranian composer and electronic musician Ash Koosha made an “AI singer” that produces surprisingly emotional lyrics. Plus how is AI changing what we see on the silver screen? The Cultural Frontline’s Laura Hubber speaks to Kelly Port, part of the team who used innovative machine learning to create the villain Thanos in the blockbuster movie, Avengers: Infinity War. Presented by Tina Daheley. Produced by Nancy Bennie, Mugabi Turya, Kirsty McQuire, Shoku Amirani, Laura Hubber and Will Coley. Image: 79530 Self-Portraits (2018) by artist Mario Klingemann on display at the Gradient Descent exhibition at Nature Morte. Credit: SV Photographic, New Delhi, courtesy of Nature Morte, New Delhi.
          Senior Site Reliability Engineer - Sift Science - Seattle, WA      Cache   Translate Page      
The Sift Science Trust PlatformTM uses real-time machine learning to accurately predict which users businesses can trust, and which ones they can't....
From Sift Science - Sun, 21 Oct 2018 06:15:49 GMT - View all Seattle, WA jobs
          Senior Software Development Engineer - Distributed Computing Services (Hex) - Amazon.com - Seattle, WA      Cache   Translate Page      
Knowledge and experience with machine learning technologies. We enable Amazon’s internal developers to improve time-to-market by allowing them to simply launch...
From Amazon.com - Thu, 26 Jul 2018 19:20:25 GMT - View all Seattle, WA jobs
          Machine Learning Engineer - Stefanini - McLean, VA      Cache   Translate Page      
AWS, Spark, Scala, Python, Airflow, EMR, Redshift, Athena, Snowflake, ECS, DevOps Automation, Integration, Docker, Build and Deployment Tools Ability to provide...
From Indeed - Tue, 16 Oct 2018 20:59:09 GMT - View all McLean, VA jobs
          Product Manager, Data - Upserve - Providence, RI      Cache   Translate Page      
Some of the technologies in our stack include React / React Native, Kinesis, DynamoDB, Machine Learning, RedShift, RDS, Ruby, JavaScript, Docker, ECS, and more....
From Upserve - Wed, 10 Oct 2018 18:03:02 GMT - View all Providence, RI jobs
          Python Developer - MJDP Resources, LLC - Wayne, PA      Cache   Translate Page      
Assemble large, complex data sets that meet business requirements and power machine learning algorithms. EC2, Lambda, ECS, S3.... $50 - $60 an hour
From Indeed - Fri, 02 Nov 2018 15:41:28 GMT - View all Wayne, PA jobs
          Executive Director- Machine Learning & Big Data - JP Morgan Chase - Jersey City, NJ      Cache   Translate Page      
We would be partnering very closely with individual lines of business to build these solutions to run on either the internal and public cloud....
From JPMorgan Chase - Thu, 01 Nov 2018 11:32:47 GMT - View all Jersey City, NJ jobs
          Sr Data Scientist Engineer (HCE) - Honeywell - Atlanta, GA      Cache   Translate Page      
50 Machine Learning. Develop relationships with business team members by being proactive, displaying a thorough understanding of the business processes and by...
From Honeywell - Thu, 20 Sep 2018 02:59:11 GMT - View all Atlanta, GA jobs
          C/C++ Software Engineer/Machine Learning - Mobica - Warszawa, mazowieckie      Cache   Translate Page      
I hereby agree for processing of personal data by Mobica Limited with headquarter in Crown House, Manchester Road, Wilmslow, UK, SK9 1BH whose representative is...
Od Mobica - Fri, 19 Oct 2018 09:05:38 GMT - Pokaż wszystkie Warszawa, mazowieckie oferty pracy
          Senior C/C++ Software Engineer/Machine Learning - Mobica - Warszawa, mazowieckie      Cache   Translate Page      
I hereby agree for processing of personal data by Mobica Limited with headquarter in Crown House, Manchester Road, Wilmslow, UK, SK9 1BH whose representative is...
Od Mobica - Fri, 14 Sep 2018 21:05:32 GMT - Pokaż wszystkie Warszawa, mazowieckie oferty pracy
          Machine Learning Researcher - PARC, a Xerox company - Palo Alto, CA      Cache   Translate Page      
PARC, a Xerox company, is in the Business of Breakthroughs®. We create new business options, accelerate time to market, augment internal capabilities, and...
From PARC, a Xerox company - Sun, 26 Aug 2018 12:12:32 GMT - View all Palo Alto, CA jobs
          Interactive Machine Learning Researcher - PARC, a Xerox company - Palo Alto, CA      Cache   Translate Page      
PARC, a Xerox company, is in the Business of Breakthroughs®. We create new business options, accelerate time to market, augment internal capabilities, and...
From PARC, a Xerox company - Sun, 26 Aug 2018 12:12:32 GMT - View all Palo Alto, CA jobs
          vScaler Cloud Adopts RAPIDS Open Source Software for Accelerated Data Science      Cache   Translate Page      

vScaler has incorporated NVIDIA’s new RAPIDS open source software into its cloud platform for on-premise, hybrid, and multi-cloud environments. Deployable via its own Docker container in the vScaler Cloud management portal, the RAPIDS suite of software libraries gives users the freedom to execute end-to-end data science and analytics pipelines entirely on GPUs. "The new RAPIDS library offers Python interfaces which will leverage the NVIDIA CUDA platform for acceleration across one or multiple GPUs. RAPIDS also focuses on common data preparation tasks for analytics and data science. This includes a familiar DataFrame API that integrates with a variety of machine learning algorithms for end-to-end pipeline accelerations without paying typical serialization costs. RAPIDS also includes support for multi-node, multi-GPU deployments, enabling vastly accelerated processing and training on much larger dataset sizes."

The post vScaler Cloud Adopts RAPIDS Open Source Software for Accelerated Data Science appeared first on insideHPC.


          Sr/Principal Consultant, Red Team - Cylance, Inc. - Texas      Cache   Translate Page      
Internal / External / Wireless - Penetration Testing (2+ years REQUIRED). By successfully applying artificial intelligence and machine learning to crack the DNA...
From Cylance, Inc. - Wed, 05 Sep 2018 19:27:50 GMT - View all Texas jobs
          PKI Engineer - Cylance, Inc. - Texas      Cache   Translate Page      
Data exchanges with internal and external security intelligence platforms. By successfully applying artificial intelligence and machine learning to crack the...
From Cylance, Inc. - Wed, 05 Sep 2018 19:27:49 GMT - View all Texas jobs
          Red Team - Consultant - Cylance, Inc. - North Carolina      Cache   Translate Page      
The consultant utilization requirements will be based on business needs. By successfully applying artificial intelligence and machine learning to crack the DNA...
From Cylance, Inc. - Tue, 23 Oct 2018 19:28:10 GMT - View all North Carolina jobs
          Technical Account Manager (East Coast, South East, Midwest - REMOTE) - Cylance, Inc. - Florida      Cache   Translate Page      
By successfully applying artificial intelligence and machine learning to crack the DNA of malware, Cylance has redefined the endpoint protection market,...
From Cylance, Inc. - Fri, 27 Jul 2018 19:28:10 GMT - View all Florida jobs
          Senior Compliance Analyst - Cylance, Inc. - Irvine, CA      Cache   Translate Page      
By successfully applying artificial intelligence and machine learning to crack the DNA of malware, Cylance has redefined the endpoint protection market,...
From Cylance, Inc. - Thu, 13 Sep 2018 19:27:37 GMT - View all Irvine, CA jobs
          Senior Compliance & Privacy Analyst - Cylance, Inc. - Irvine, CA      Cache   Translate Page      
By successfully applying artificial intelligence and machine learning to crack the DNA of malware, Cylance has redefined the endpoint protection market,...
From Cylance, Inc. - Sat, 08 Sep 2018 01:27:53 GMT - View all Irvine, CA jobs
          Financial Reporting Director - Cylance, Inc. - Irvine, CA      Cache   Translate Page      
By successfully applying artificial intelligence and machine learning to crack the DNA of malware, Cylance has redefined the endpoint protection market,...
From Cylance, Inc. - Mon, 13 Aug 2018 07:27:33 GMT - View all Irvine, CA jobs
          Services Operations Manager - Highspot - Seattle, WA      Cache   Translate Page      
Equipped with new Apple products. We employ advanced technologies, included patented machine learning algorithms to:....
From Highspot - Mon, 01 Oct 2018 20:07:27 GMT - View all Seattle, WA jobs
          Mobile Engineer React Native - Xevo - Bellevue, WA      Cache   Translate Page      
Technical and non-technical, internal and external. Do you dream about artificial intelligence and machine learning?...
From Xevo - Tue, 02 Oct 2018 02:28:59 GMT - View all Bellevue, WA jobs
          Financial Data Science Analyst - Apple - Austin, TX      Cache   Translate Page      
Our team applies data science and machine learning to drive strategic impact across multiple lines of business at Apple....
From Apple - Wed, 07 Nov 2018 01:45:42 GMT - View all Austin, TX jobs
          Full Stack Engineer - Hired - San Francisco, CA      Cache   Translate Page      
The company is backed by Lumia Capital, Sierra Ventures, and other leading investors. Work with Machine Learning &amp; Search engineers to create a powerful yet...
From Hired - Wed, 17 Oct 2018 00:24:01 GMT - View all San Francisco, CA jobs
          Senior Software Engineer, Search - Hired - San Francisco, CA      Cache   Translate Page      
The company is backed by Lumia Capital, Sierra Ventures, and other leading investors. Desire to learn about how Data &amp; Machine Learning can influence Search &amp;...
From Hired - Wed, 08 Aug 2018 22:32:04 GMT - View all San Francisco, CA jobs
          Senior Machine Learning Engineer - Hired - San Francisco, CA      Cache   Translate Page      
Have strong understanding of product, business, key KPIs across product and business. Hired is looking for a Senior Machine Learning Engineer to join our Search...
From Hired - Tue, 31 Jul 2018 02:30:02 GMT - View all San Francisco, CA jobs
          Senior Technical Product Manager – Machine Learning - DefinedCrowd - Seattle, WA      Cache   Translate Page      
Familiarity with data storage technologies and patterns, distributed systems, data warehouse/data lake and ETL architectures with Hadoop ecosystem, such as...
From Definedcrowd - Tue, 06 Nov 2018 15:44:08 GMT - View all Seattle, WA jobs
          Solutions Architect - NVIDIA - Washington State      Cache   Translate Page      
Assist field business development in through the enablement process for GPU Computing products, technical relationship and assisting machine learning/deep...
From NVIDIA - Sun, 19 Aug 2018 07:55:46 GMT - View all Washington State jobs
          Solutions Architect - NVIDIA - Virginia      Cache   Translate Page      
Build and cultivate internal understanding of data analytics and machine learning among the NVIDIA technical community....
From NVIDIA - Tue, 02 Oct 2018 07:57:56 GMT - View all Virginia jobs
          US Federal CTO - Solution Architect - NVIDIA - Virginia      Cache   Translate Page      
Build and cultivate internal understanding of Federal customer data analytics and machine learning requirements among the NVIDIA technical community including...
From NVIDIA - Tue, 07 Aug 2018 01:54:32 GMT - View all Virginia jobs
          Fullstack Software Engineer | New York - BenevolentAI - New York, NY      Cache   Translate Page      
Machine Learning Squads. All employment is decided on the basis of qualifications, merit, and business need. Fun internal events (boat parties, karting, Oktober...
From BenevolentAI - Sat, 27 Oct 2018 10:26:13 GMT - View all New York, NY jobs
          Solutions Architect - NVIDIA - Maryland      Cache   Translate Page      
You will also be an internal champion for Data Analytics and Machine Learning among the NVIDIA technical community....
From NVIDIA - Tue, 06 Nov 2018 07:57:54 GMT - View all Maryland jobs
          Sr. Solutions Architect - Team Lead - NVIDIA - Maryland      Cache   Translate Page      
You will also be an internal champion for Data Analytics and Machine Learning among the NVIDIA technical community....
From NVIDIA - Wed, 12 Sep 2018 19:54:41 GMT - View all Maryland jobs
          Senior Solution Architect - Cyber Security - NVIDIA - Maryland      Cache   Translate Page      
You will also be an internal champion for Cyber Security and Machine Learning among the NVIDIA technical community....
From NVIDIA - Wed, 01 Aug 2018 07:57:49 GMT - View all Maryland jobs
          Sr. Solutions Architect - Lead - NVIDIA - Washington, DC      Cache   Translate Page      
You will also be an internal champion for Data Analytics and Machine Learning among the NVIDIA technical community....
From NVIDIA - Wed, 12 Sep 2018 19:54:40 GMT - View all Washington, DC jobs
          Sr Data Scientist - NVIDIA - Santa Clara, CA      Cache   Translate Page      
3+ years’ experience in solving problems using machine learning algorithms and techniques (clustering, classification, outlier analysis, etc.)....
From NVIDIA - Tue, 30 Oct 2018 01:54:49 GMT - View all Santa Clara, CA jobs
          Help me with machine learning in python      Cache   Translate Page      
I am currently doing Machine Learning in the university and I need some help understanding all the classes. I need someone with knowledge in Python as well as ML (Budget: €30 - €250 EUR, Jobs: Machine Learning, Python, Software Architecture)
          AMD Unveils World's First 7nm Datacenter GPUs with PCIe 4.02 Interconnect      Cache   Translate Page      
AMD unveiled the world's first lineup of 7nm GPUs for the datacenter that will utilize an all new version of the ROCM open software platform for accelerated computing. "The AMD Radeon Instinct MI60 and MI50 accelerators feature flexible mixed-precision capabilities, powered by high-performance compute units that expand the types of workloads these accelerators can address, including a range of HPC and deep learning applications." They are specifically designed to tackle datacenter workloads such as rapidly training complex neural networks, delivering higher levels of floating-point performance, while exhibiting greater efficiencies. The new "Vega 7nm" GPUs are also the world's first GPUs to support the PCIe 4.02 interconnect which is twice as fast as other x86 CPU-to-GPU interconnect technologies and features AMD Infinity Fabric Link GPU interconnect technology that enables GPU-to-GPU communication that is six times faster than PCIe Gen 3. The AMD Radeon Instinct MI60 Accelerator is also the world's fastest double precision PCIe accelerator with 7.4 TFLOPs of peak double precision (FP64) performance. "Google believes that open source is good for everyone," said Rajat Monga, engineering director, TensorFlow, Google. "We've seen how helpful it can be to open source machine learning technology, and we're glad to see AMD embracing it. With the ROCm open software platform, TensorFlow users will benefit from GPU acceleration and a more robust open source machine learning ecosystem." ROCm software version 2.0 provides updated math libraries for the new DLOPS; support for 64-bit Linux operating systems including CentOS, RHEL and Ubuntu; optimizations of existing components; and support for the latest versions of the most popular deep learning frameworks, including TensorFlow 1.11, PyTorch (Caffe2) and others. Discussion
          Why you should use Gandiva for Apache Arrow      Cache   Translate Page      

Over the past three years Apache Arrow has exploded in popularity across a range of different open source communities. In the Python community alone, Arrow is being downloaded more than 500,000 times a month. The Arrow project is both a specification for how to represent data in a highly efficient way for in-memory analytics, as well as a series of libraries in a dozen languages for operating on the Arrow columnar format.

In the same way that most automobile manufacturers OEM their transmissions instead of designing and building their own, Arrow provides an optimal way for projects to manage and operate on data in-memory for diverse analytical workloads, including machine learning, artificial intelligence, data frames, and SQL engines.

To read this article in full, please click here

(Insider Story)
          Solutions Engineer (JavaScript)      Cache   Translate Page      
CA-Milpitas, If you are a Solutions Engineer with a software/technical background, please read on! Based in Milpitas, near North San Jose, California - we are an early stage Series A funded for profit startup allowing real time stream processing platform to interact across all digital channels to be built from multiple digital sources. We leverage machine learning concepts to provide the best experience possib
          Lead Machine Learning Software Engineer - up to 220k + bonus      Cache   Translate Page      
CA-San Jose, Location: Work remote initially, once established office will be either Redwood City OR San Jose. Will need to be in the office 2-3 days per week minimum. Salary: Up to 220k base plus bonus Skills: Machine learning, full lifecycle experience, programming with a variety of languages Work for an industry leader which is one of the largest consumer products brands around the globe! It's an exciting t
          BXB Digital: IoT Technical Program Manager      Cache   Translate Page      
CA-Santa Clara, Job ID #: 7949 BXB Digital connects product conveyance platforms with digital capabilities to create more connected, intelligent and efficient supply chains. Established in 2016, BXB Digital is the newest business unit within the Brambles family, the global leader in supply chain logistics. BXB Digital combines network concepts, enterprise supply chain expertise, machine learning and the Internet
          Comentário sobre Como o Brasil se transformou em terreno fértil para a difusão de notícias falsas por Luiz Carlos Feliponi      Cache   Translate Page      
Somente inteligencia artificial pode resolver o problema. Um IA poderia, através de palavras chaves dentro da mensagem, realizar buscas em veículos renomados de noticias e enviar um alerta para o usuario, avisando-o que a notifica "pode" ser falsa. Nao é algo dificil de implementar e acho que o Facebook está fazendo algo parecido com Machine Learning.
          Sr. Associate, Data Modeler, Financial Services - KPMG - Seattle, WA      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 07 Sep 2018 02:02:14 GMT - View all Seattle, WA jobs
          Sr. Associate, ML Pipelines for AI Consultant - KPMG - McLean, VA      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 26 Oct 2018 15:22:01 GMT - View all McLean, VA jobs
          Sr. Associate, AI in Management Analytics Consultant - KPMG - McLean, VA      Cache   Translate Page      
Ability to apply statistical, machine learnings, and artificial intelligence techniques to achieve concrete business goals and work with the business to...
From KPMG LLP - Sat, 29 Sep 2018 15:21:53 GMT - View all McLean, VA jobs
          Data Scientist - Deloitte - Springfield, VA      Cache   Translate Page      
Demonstrated knowledge of machine learning techniques and algorithms. We believe that business has the power to inspire and transform....
From Deloitte - Fri, 10 Aug 2018 06:29:44 GMT - View all Springfield, VA jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Dallas, TX      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 07 Sep 2018 02:02:14 GMT - View all Dallas, TX jobs
          Associate, Machine Learning AI Consultant, Financial Services - KPMG - Dallas, TX      Cache   Translate Page      
Broad, versatile knowledge of analytics and data science landscape, combined with strong business consulting acumen, enabling the identification, design and...
From KPMG LLP - Fri, 07 Sep 2018 02:02:14 GMT - View all Dallas, TX jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - Philadelphia, PA      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 14 Sep 2018 08:38:34 GMT - View all Philadelphia, PA jobs
          Event Scientist - KeyBank - Brooklyn, OH      Cache   Translate Page      
ABOUT THE BUSINESS:. Knowledge of machine learning models. Research and assess available technologies in relation to data analytics, reporting, visualization...
From KeyBank - Tue, 06 Nov 2018 22:09:57 GMT - View all Brooklyn, OH jobs
          Sr. Associate, Data Modeler, Financial Services - KPMG - New York, NY      Cache   Translate Page      
Modeling (regression, machine learning, feature selection, dimension reduction, validation); Strong aptitude for quickly learning business operational, process,...
From KPMG LLP - Fri, 14 Sep 2018 08:38:34 GMT - View all New York, NY jobs
          Associate, Machine Learning AI Consultant, Financial Services - KPMG - New York, NY      Cache   Translate Page      
Broad, versatile knowledge of analytics and data science landscape, combined with strong business consulting acumen, enabling the identification, design and...
From KPMG LLP - Fri, 14 Sep 2018 08:38:34 GMT - View all New York, NY jobs
          Data Scientist: Medical VoC and Text Analytics Manager - GlaxoSmithKline - Research Triangle Park, NC      Cache   Translate Page      
Strong business acumen; 2+ years of unstructured data analysis/text analytics/natural language processing and/or machine learning application for critical...
From GlaxoSmithKline - Fri, 19 Oct 2018 23:19:12 GMT - View all Research Triangle Park, NC jobs
          Linux Systems Administrator - Kira Systems - Toronto, ON      Cache   Translate Page      
Growth potential, values and attitudes are equally important to be a successful Kiran. Kira Systems is a powerful machine learning software that identifies,...
From Indeed - Tue, 06 Nov 2018 16:04:32 GMT - View all Toronto, ON jobs
          Linear Regression – Machine Learning with TensorFlow and Oracle JET UI Explained by Andrejus ...      Cache   Translate Page      
image

Machine learning topic is definitely popular these days. Some get wrong assumptions about it - they think machine could learn by itself and its kind of magic. The truth is - there is no magic, but math behind it. Machine will learn the way math model is defined for learning process. In my opinion, the best solution is a combination of machine learning math and algorithms.  Here I could relate to chatbots keeping conversational context - language processing can be done by machine learning with neural network, while intent and context processing can be executed by programmable algorithms.
If you are starting to learn machine learning - there are two essential concepts to start with:
1. Regression
2. Classification
This post is focused around regression, in the next posts I will talk about classification.
Regression is a method which calculates the best fit for a curve to summarize data. Its up to you which type of curve to choose, you should assume which type will be most suitable (this can be achieved with trial and error too) based on given data set. Regression goal is to understand data points by discovering the curve that might have generated them. Read the complete article here.

 

Developer Partner Community

For regular information become a member in the WebLogic Partner Community please visit: http://www.oracle.com/partners/goto/wls-emea ( OPN account required). If you need support with your account please contact the Oracle Partner Business Center.

Blog Twitter LinkedIn Forum Wiki

Technorati Tags: PaaS,Cloud,Middleware Update,WebLogic, WebLogic Community,Oracle,OPN,Jürgen Kress


          AI: Researchers use machine learning to find source of viruses      Cache   Translate Page      

AI in healthcareScientists have developed a machine learning algorithm that may help find the original hosts of viruses. It is hoped that the new tool could help inform preventive measures against deadly diseases. The new research, led by the University of Glasgow, uses a new algorithm designed to use viral genome sequences to predict the likely natural […]

The post AI: Researchers use machine learning to find source of viruses appeared first on Internet of Business.


          Chief Architect (Data Analytics - Strategic Technology Centre) - Singapore Technologies Engineering Ltd - Ang Mo Kio      Cache   Translate Page      
Location based services using Google Map, ESRI, Trillium etc. is a plus. ST Engg has identified data analytics, machine learning and Artificial Intelligence as...
From Singapore Technologies Electronics - Fri, 12 Oct 2018 06:27:13 GMT - View all Ang Mo Kio jobs
          Designing machine learning — Research Imagining      Cache   Translate Page      
The relationship between user experience (UX) designers and machine learning (ML) data scientists has emerged as a site for research since 2017. Central to recent findings is the limited ability of UX designers to conceive of new ways to use ML (Yang et al. 2018). This is due to a number of factors. Firstly, human […] […]
          The #AIEYE: Gopher's (OTCQB: $GOPH) New Version of Avant! #AI, IBM (NYSE: $IBM) #Watson Health Publishes Top 50 Cardiovascular Hospitals and ABB Optical Tech (NYSE: $ABB) Launched Onboard Japanese Satellite      Cache   Translate Page      
Point Roberts, WA and Vancouver, BC - November 6, 2018 (Investorideas.com Newswire) Investorideas.com (www.investorideas.com), a global investor news source covering Artificial Intelligence (AI) , in partnership with Gopher Protocol's (OTCQB: GOPH). Avant!, the latest innovation in machine learning, brings you today's edition of The AI Eye - Watching stock news, deal tracker and advancements in artificial intelligence.
          The #AIEye: Gopher's (OTCQB: $GOPH) Avant! AI Moves Toward Personalization, Accenture (NYSE: $ACN) to Open R&D Lab in Shenzhen, China and Fortinet Expands OT Security Leadership      Cache   Translate Page      
Point Roberts, WA and Vancouver, BC - November 5, 2018 (Investorideas.com Newswire) Investorideas.com (www.investorideas.com), a global investor news source covering Artificial Intelligence (AI) , in partnership with Gopher Protocol's (OTCQB: GOPH). Avant! , the latest innovation in machine learning, brings you today's edition of The AI Eye - Watching stock news, deal tracker and advancements in artificial intelligence.
          The Bridge Limited: Data Scientist / Machine Learning Engineer      Cache   Translate Page      
The Bridge Limited: Data Scientist / Machine Learning Engineer Fantastic opportunity to join a leading UK client on a long term contract. Our client requires a Data Scientist to join them and to be able to hit the ground running. The skills required for this role: Machine Le Hatfield
          Look out for the cyber threats hiding in your backups      Cache   Translate Page      

Look out for the cyber threats hiding in your backups

Spending on security technology continues to soar. Nevertheless, data breaches and cyberattacks continue to make headlines at an incredible rate, with no relief in sight. The Online Trust Alliance reported that attacks in 2017 came from a myriad of vectors, such as phishing and ransomware, and that the number of attacks doubled to nearly 160,000 incidents per year over 2016. What’s worse, estimates for the number of unreported attacks exceed 350,000 annually.

While enterprises typically dominate the headlines, organizations of all sizes are affected by cyber incidents. A recent Ponemon study showed that two-thirds of small and mid-sized businesses reported that threats evaded their intrusion detection systems, and more than half of the companies said they were attacked by ransomware more than twice during the last year. There is no dispute that the number of vulnerable endpoints and the complexity of threats will continue to increase, and limited IT budgets and overstretched staff will remain an industry-wide problem. It’s clear that companies need to adopt new approaches to stay ahead of cyberattacks.

Traditional approaches aren’t enough

Firewalls and antivirus solutions are the norm in most IT shops, and they do thwart security attacks daily. Despite being very widely deployed, industry trends clearly show the need for more innovative approaches to threat detection.

Larger organizations have the resources to implement security incident & event management (SIEM) solutions, which are effective in collecting vast amounts of data from endpoints. Arrays of sensors can be integrated to provide rich and comprehensive security data for many types of cyber analytics. But big data platforms are complex to deploy and manage, and even the most advanced IT shops admit to not being able to keep up with the sheer volume of flags and false positives. Unfortunately, hidden in the sea of alerts are numerous threats that do get through, untriaged and undetected.

Whether or not a company has the resources for the most advanced cybersecurity tools and specialized personnel to support them, the fact remains that traditional solutions are not stopping threats from compromising critical network systems.

Gaps in the armor

Backup and disaster recovery systems are the go-to resources as insurance policies to protect against cyberattacks. Successful recovery from the last known good configuration is a reasonable and sound approach, but this assumes knowing exactly when the attack occurred, as well as discipline with ongoing backup testing. A 2018 benchmark study sponsored by IBM revealed that a mean time to identify a data breach incident climbed to a staggering 197 days with another 69 days to contain it. Backups are being contaminated during this lengthy timeframe. Thus, the recovery process following a cyber incident will not only be highly labor-intensive but also a protracted affair, if it is even possible at all.

Compounding matters, breaches rarely remain confined and often spread across a corporate network, compromising a variety of systems and databases, making the recovery effort even more complex. As malware continues to increase in sophistication, the level of manual effort required to unravel the intricate cybersecurity maze and restore all system components of a production environment can be immense.

Looking forward

In the future, we will see intelligent and automated tools that use granular backup and replication data sets to continuously detect latent security breaches, irregular behavior and patterns or other unusual backup attributes that may pose a risk to a quick recovery. Offline backup and replication files and their metadata are an untapped wealth of context-rich cybersecurity data, which opens a new door to proactively identifying cyber threats without impacting production workloads. With the right tools, organizations of all sizes could leverage this data to improve their security posture and decrease the labor-intensive manual effort currently required in recovery efforts.

Soon, security fingerprinting, data transformation, machine learning and advanced analytics will all be brought to bear to automatically analyze backup and replication data for cybersecurity issues in near real-time. By integrating with threat intelligence feeds, verified cyber infections can be readily detected and confirmed. Malware and other anomalous behavior can be surfaced and resolved long before the backups or replicas are needed following a disaster. These next-generation cyber tools will enable data protection systems to deliver the speedy recovery that’s both needed and expected.

Compared to current solutions, this new approach will dramatically improve security analysis and remediation, since backup data is offline, every restore point can be tested automatically and analysis does not impact production environments. Utilizing backup and replication data sets for cybersecurity purposes is a completely new methodology for attaining a multi-layered security program at a cost that’s practical for all IT shops, not just enterprises with enormous budgets.

New approaches can provide dramatic improvements in the security posture of an organization by examining a previously inaccessible, but endlessly rich data source for advance threat detection, analysis and remediation. It’s widely agreed that every organization would benefit from more effective cybersecurity tools. Forward-looking IT shops will begin turning to the very backup and replication data sets already being captured to create a more robust, cost-effective cybersecurity strategy.

Photo Credit: lolloj / Shutterstock


Look out for the cyber threats hiding in your backups
Lynn LeBlanc, CEO and founder of HotLink Corporation , has more than 25 years of enterprise software and technology experience at both Fortune 500 companies and Silicon Valley startups. Prior to founding HotLink, Lynn was founder and CEO of FastScale Technology, an enterprise software company acquired by VMware, Inc.
          Saudi Stock Market Analysis and Forecasting (Tadawul)―Part I      Cache   Translate Page      

Saudi Stock Market Analysis and Forecasting (Tadawul)―Part I

This is my first project in financial industry, especially after spending a lot of time conducting machine learning projects in healthcare sectors (such as A&E Attendances and Emergency Admissions in England , and Breast Cancer Cell Type Classifier ). So, I have decided to utilize my skills in one of the biggest stock markets in the middle east, which is Saudi Arabia Stock Market (Tadawul). Although, data acquiring and collection is a challenging part. But thanks to Tadawul that made it quite easy! There I had to go through all Saudi Market Activities annual reports that published on their website, extract the required data and finally consolidate them in one CSV file. In the end, I had a dataset of Saudi Arabia Market Activities starting from Jan 2006 until Aug 2018.

More about Tadawul : Tadawul is the sole entity authorized in the Kingdom of Saudi Arabia to act as the Securities Exchange (the Exchange). It mainly carries out listing and trading in securities, as well as deposit, transfer, clearing, settlement, and registry of ownership of securities traded on the Exchange”

This project aims to analyze Saudi Arabia stock market data starting from 2006 until Aug-2018 and make prediction/forecasting for 12 months (Sep-2018 until Aug-2019). I will be accomplishing this project in a series, where each part will be discussing in details different topics.

PART I ― Loading Dataset, Data Wrangling, Data Analysis & Exploration Loading Dataset

Firstly, we load the required libraries then dataset:

library(dplyr) library(urca) library(forecast) library(repr) library(data.table) library(psych) library(dygraphs) require(ggplot2)

Here is a screenshot of the consolidated Monthly Market Activities dataset that I had consolidated from Tadawul reports and publications ( 4 X 152 ). All currency units are in Saudi Riyal (SAR):


Saudi Stock Market Analysis and Forecasting (Tadawul)―Part I

Loading dataset:

df <- read.csv ( file= "… /Tadawul/Datasets/Monthly Market Activity.csv", header= TRUE ) head(df)
Saudi Stock Market Analysis and Forecasting (Tadawul)―Part I

Changing header names as the followings:

Month: Date

Value.of.Shares.Traded (SAR): VST

Number.of.Shares.Traded: NST

Transactions: Tra

setnames(df, old=c("Month","Value.of.Shares.Traded","Number.of.Shares.Traded", "Transactions"), new=c("Date", "VST", "NST","Tra")) head(df)
Saudi Stock Market Analysis and Forecasting (Tadawul)―Part I
Data Wrangling

Checking data types:

class(df$Date) class(df$VST) class(df$NST) class(df$Tra) > 'factor' > 'numeric' > 'numeric' > 'numeric'

The class of column ‘Date’ is ‘factor’. For time series analysis, we should have date and time data type. But first, let us check if there is any missing data exists:

summary(is.na(df$Date)) summary(is.na(df$VST)) summary(is.na(df$NST)) summary(is.na(df$Tra)) > Mode FALSE
> logical 152
>
> Mode FALSE
> logical 152
>
> Mode FALSE
> logical 152
>
> Mode FALSE
> logical 152

No missing data, changing ‘Date’ column type into date and time class ‘POSIXct’ ‘POSIXt’:

to.POSIXct <- function(col){ dateStr <- paste(as.character(col)) as.POSIXct( strptime(dateStr, "%d/%m/%Y")) } df$Date <- to.POSIXct(df$Date) class(df$Date) > 'POSIXct' 'POSIXt'

Create a ‘month’ column, it helps with Data Visualization and Exploration:

df$month = month(df$Date) df$month = month.abb[df$month] head(df)
Saudi Stock Market Analysis and Forecasting (Tadawul)―Part I
Data Analysis, Visualization, and Exploration

Let us check the Data-frame summary first:

summary(df[,2:5]) > VST NST Tra month > Min. :3.240e+10 Min. :1.557e+09 Min. : 3185 Length:152 > 1st Qu.:8.006e+10 1st Qu.:3.503e+09 1st Qu.: 1969411 Class :character > Median :1.163e+11 Median :4.669e+09 Median : 2707780 Mode :character > Mean :1.487e+11 Mean :4.875e+09 Mean : 3254791 > 3rd Qu.:1.731e+11 3rd Qu.:6.125e+09 3rd Qu.: 3787704 Max. :8.284e+11 Max. :1.383e+10 Max. :12105358

Scatter plot of Value of Shares Traded (VST) in the time period (Jan 2006 ― Aug 2018), color-coded with Number of shares Traded (NST):

ggplot(df, aes(Date, VST, color = NST)) + geom_point(shape = 16, size = 2, show.legend = TRUE) + theme_minimal()
Saudi Stock Market Analysis and Forecasting (Tadawul)―Part I

Checking the data distribution (histogram plots) and the relationship between the value of shares traded, the number of shares traded and number of Transactions:

pairs.panels(df[,1:3], method = "pearson", # correlation method hist.col = "#00AFBB", density = TRUE, # show density plots ellipses = TRUE # show correlation ellipses)
Saudi Stock Market Analysis and Forecasting (Tadawul)―Part I

Clearly, there are outliers or big numbers in VST, NST, and Tra that skews the distribution curves, mostly those number at 2006 2008. We will see exactly where are those numbers in time series plotting and analysis. However, there is a clear linearity between the value of shares traded (VST) and number transactions (Tra). So, a higher number of transactions means higher value of the shares traded. Pearson correlation method can positively approve that by showing the strength of a linear relationship between VST and Tra (0.89).

Indeed, I am interested to learn more about the Average Value of each share traded during the time period between (2006 2018). But first, we have to create this feature which will name it (Avg.V) by dividing VST over NST:

# Average value of each share Avg.V
df$Avg.V <- df$VST/df$NST
summary(df$Avg.V) > Min. 1st Qu. Median Mean 3rd Qu. Max. > 13.97 21.25 24.53 29.60 32.39 104.78

Checking the average value for each share distribution:

ggplot(df, aes(x =Avg.V)) + geom_histogram()
Saudi Stock Market Analysis and Forecasting (Tadawul)―Part I

Scatter plot for ‘Avg.V’ against ‘Date’ with ‘NST’ color coding:

ggplot(df, aes(Date, Avg.V, color = NST)) + geom_point(shape = 16, size = 2, show.legend = TRUE) + theme_minimal()
          Ingénieur Machine Learning en imagerie      Cache   Translate Page      
France - supérieure, (Ecole d'ingénieur ou Université Bac +5) vous avez un doctorat dans le domaine de la reconnaissance de formes, apprentissage (machine...
          Micron's bet: Quad-level cell NAND SSDs will finally replace HDDs      Cache   Translate Page      
The company's Micron 5210 ION enterprise SATA SSD is now generally available and aimed at artificial intelligence, machine learning, deep learning and other intensive workloads.

           Windows-Based bandwidth allocation on optical networks       Cache   Translate Page      
Mahadevan, V. and Yu, W. and Zhou, J. (2009) Windows-Based bandwidth allocation on optical networks. In: International Conference on Machine Learning and Computing (IACSIT ICMLC 2009), JUL 10-12, 2009 , Perth, AUSTRALIA.
           Router-Based bandwidth allocation on optical networks       Cache   Translate Page      
Mahadevan, V. and Yu, W. and Zhou, J. (2009) Router-Based bandwidth allocation on optical networks. In: International Conference on Machine Learning and Computing (IACSIT ICMLC 2009), JUL 10-12, 2009 , Perth, AUSTRALIA.
           Static bandwidth allocation on optical networks       Cache   Translate Page      
Mahadevan, V. and Yu, W. and Zhou, J. (2009) Static bandwidth allocation on optical networks. In: International Conference on Machine Learning and Computing (IACSIT ICMLC 2009, JUL 10-12, 2009 , Perth, AUSTRALIA.
          Machine Learning Engineer / Algorithm Developer - TECHNICA CORPORATION - Dulles, VA      Cache   Translate Page      
Job Description: We are seeking a highly creative software engineer experienced in artificial intelligence and deep learning techniques to design, develop,...
From Technica Corporation - Fri, 05 Oct 2018 10:31:19 GMT - View all Dulles, VA jobs
          UR - Corporate Engineering - Precision Systems Engineer (Maplewood, MN) - 3M - Maplewood, MN      Cache   Translate Page      
Proactively collaborate with business partners to connect and extend process data management solutions with complimentary machine learning and analytics efforts...
From 3M - Wed, 05 Sep 2018 17:09:45 GMT - View all Maplewood, MN jobs
          Data Science Manager - Micron - Boise, ID      Cache   Translate Page      
Create server based visualization applications that use machine learning and predictive analytic to bring new insights and solution to the business....
From Micron - Wed, 05 Sep 2018 11:18:49 GMT - View all Boise, ID jobs
          Intern - Data Scientist (NAND) - Micron - Boise, ID      Cache   Translate Page      
Machine learning and other advanced analytical methods. To ensure our software meets Micron's internal standards....
From Micron - Wed, 29 Aug 2018 20:54:50 GMT - View all Boise, ID jobs
          Intern - Data Scientist (DRAM) - Micron - Boise, ID      Cache   Translate Page      
Machine learning and other advanced analytical methods. To ensure our software meets Micron's internal standards....
From Micron - Mon, 20 Aug 2018 20:48:37 GMT - View all Boise, ID jobs
          Machine Learning Engineer - TECHNICA CORPORATION - Dulles, VA      Cache   Translate Page      
Technica Corporation is seeking a Machine Learning. Engineer to support our internal Innovation, Research....
From Technica Corporation - Thu, 23 Aug 2018 10:27:08 GMT - View all Dulles, VA jobs
          Director of Artificial Intelligence - Micron - Milpitas, CA      Cache   Translate Page      
Broad, versatile knowledge of artificial intelligence and machine learning landscape, combined with strong business consulting acumen, enabling the...
From Micron - Fri, 19 Oct 2018 17:30:05 GMT - View all Milpitas, CA jobs
          Product Operations Intern - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 17 Oct 2018 22:29:11 GMT - View all New York, NY jobs
          QA Engineer - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Sun, 07 Oct 2018 06:47:33 GMT - View all New York, NY jobs
          Data Scientist - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Thu, 04 Oct 2018 06:17:29 GMT - View all New York, NY jobs
          NLP Data Scientist - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Thu, 27 Sep 2018 14:36:48 GMT - View all New York, NY jobs
          Business Development Representative - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 26 Sep 2018 16:31:17 GMT - View all New York, NY jobs
          UI/UX Designer - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 26 Sep 2018 16:18:14 GMT - View all New York, NY jobs
          Customer Success Manager - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 26 Sep 2018 14:35:24 GMT - View all New York, NY jobs
          Software Architect - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 26 Sep 2018 14:35:24 GMT - View all New York, NY jobs
          UI Engineer - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Tue, 25 Sep 2018 06:27:49 GMT - View all New York, NY jobs
          Mid-Market Account Executive - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Mon, 24 Sep 2018 16:34:54 GMT - View all New York, NY jobs
          Full Stack Engineer - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Jazz VP, Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make...
From Wade & Wendy - Wed, 19 Sep 2018 06:16:43 GMT - View all New York, NY jobs
          Enterprise Account Executive - Wade & Wendy - New York, NY      Cache   Translate Page      
Our team is backed by Slack, ffVC, Randstad and other great VCs, as we bring AI and machine learning to the recruiting/HR space - all in order to make the...
From Wade & Wendy - Mon, 10 Sep 2018 14:31:19 GMT - View all New York, NY jobs
          Senior Java Engineer - Java, Spring, AngularJS      Cache   Translate Page      
CA-Irvine, If you are a Senior Java Engineer with experience, please read on! Top Reasons to Work with Us CyberCoders is a very technology enabled recruitment firm that leverages best in class technology to match candidates with our clients. We use big data, machine learning, and artificial intelligence to build tools & systems for our recruitment teams. What You Will Be Doing This role is for you is a Sr. J
          Director, Marketing Automation for Relationship Marketing - Microsoft - Redmond, WA      Cache   Translate Page      
In addition, this role will lead the identification and leveraging of the latest marketing capabilities like machine learning and AI to propel Relationship...
From Microsoft - Wed, 31 Oct 2018 07:57:36 GMT - View all Redmond, WA jobs
          Order Processing Specialist - Pure Storage - Salt Lake City, UT      Cache   Translate Page      
The world is experiencing a technological revolution driven by AI, machine learning, virtual reality, quantum computing and self-driving cars -- all of which...
From Pure Storage - Tue, 28 Aug 2018 06:24:19 GMT - View all Salt Lake City, UT jobs
          Travel and Expense Specialist - Pure Storage - Salt Lake City, UT      Cache   Translate Page      
Bachelor's degree in business, finance and accounting related field. The world is experiencing a technological revolution driven by AI, machine learning,...
From Pure Storage - Tue, 28 Aug 2018 06:23:33 GMT - View all Salt Lake City, UT jobs
          Technical Support Engineer II - FlashArray (Second Shift) - Pure Storage - Lehi, UT      Cache   Translate Page      
The world is experiencing a technological revolution encouraged by AI, machine learning, virtual reality, quantum computing and self-driving cars -- all of...
From Pure Storage - Fri, 19 Oct 2018 22:33:21 GMT - View all Lehi, UT jobs
          Technical Support Engineer II/III - Pure Storage - Lehi, UT      Cache   Translate Page      
The world is experiencing a technological revolution driven by AI, machine learning, virtual reality, quantum computing and self-driving cars -- all of which...
From Pure Storage - Sun, 07 Oct 2018 06:44:59 GMT - View all Lehi, UT jobs
          Technical Support Engineer II - NAS/ Storage - Pure Storage - Lehi, UT      Cache   Translate Page      
The world is experiencing a technological revolution driven by AI, machine learning, virtual reality, quantum computing and self-driving cars -- all of which...
From Pure Storage - Wed, 01 Aug 2018 06:20:10 GMT - View all Lehi, UT jobs
          Solution Architect - Data & Analytics - Neudesic LLC - New York, NY      Cache   Translate Page      
Machine Learning Solutions:. The explosion of big data, machine learning and cloud computing power creates an opportunity to make a quantum leap forward in...
From Neudesic LLC - Mon, 15 Oct 2018 09:58:30 GMT - View all New York, NY jobs
          Enterprise Pursuit Manager - Pure Storage - Boston, MA      Cache   Translate Page      
The world is experiencing a technological revolution driven by AI, machine learning, virtual reality, quantum computing and self-driving cars -- all of which...
From Pure Storage - Wed, 03 Oct 2018 22:31:54 GMT - View all Boston, MA jobs
          Machine Learning Researcher - PARC, a Xerox company - Palo Alto, CA      Cache   Translate Page      
PARC, a Xerox company, is in the Business of Breakthroughs®. We create new business options, accelerate time to market, augment internal capabilities, and...
From PARC, a Xerox company - Sun, 26 Aug 2018 12:12:32 GMT - View all Palo Alto, CA jobs
          Interactive Machine Learning Researcher - PARC, a Xerox company - Palo Alto, CA      Cache   Translate Page      
PARC, a Xerox company, is in the Business of Breakthroughs®. We create new business options, accelerate time to market, augment internal capabilities, and...
From PARC, a Xerox company - Sun, 26 Aug 2018 12:12:32 GMT - View all Palo Alto, CA jobs
          Computer Science and Operations Research - Three (3) Faculty Positions (Machine Learning) - Université de Montréal - Québec City, QC      Cache   Translate Page      
In connection with the grant of the Canada First Research Excellence Fund (IVADO), the Department of Computer Science and Operations Research is seeking...
From University Affairs - Thu, 11 Oct 2018 18:28:23 GMT - View all Québec City, QC jobs
          AI & Machine Learning: Matching talent with the right job at the right time      Cache   Translate Page      

Artificial Intelligence (AI) and machine learning are changing the world of talent acquisition. While talent is still scarce, it is now easier and quicker than ever to identify the right candidates for a job. So how do AI and machine learning help recruiters to be more efficient at finding and connecting with the right talent at the right time?


          Machine learning: hype of here to stay?      Cache   Translate Page      
Weten we eigenlijk wel waar we het over hebben als we iets betitelen als machine learning of data driven? Doorvragen leidt tot complexe antwoorden. Het begrijpen is juist vrij essentieel in het bepalen of we het nu over een hype hebben of een techniek die de maatschappij en ons leven aanzienlijk verandert. De term machine […]
          Architect - Microsoft - Redmond, WA      Cache   Translate Page      
Preferred experience with at least one of the Machine Learning related technologies (SAS, SPSS, RevR, Azure ML, MapR)....
From Microsoft - Thu, 23 Aug 2018 08:25:24 GMT - View all Redmond, WA jobs
          Cardinal Analytx Bounds into Next Growth Phase with Former DecisionView CEO Linda Hand at Helm      Cache   Translate Page      

PALO ALTO, Calif., Nov. 7, 2018 /PRNewswire/ -- Cardinal Analytx Solutions, a developer of advanced machine learning solutions for improving healthcare while reducing costs by predicting impending medical conditions, today announced it has hired healthcare veteran Linda T. Hand as CEO to...


          MACHINE LEARNING INTERN FOR SPEECH RELATED APPLICATIONS - Huawei Canada - Montréal, QC      Cache   Translate Page      
We thank all applicants for their interest in career opportunities with Huawei. Machine Learning Intern for Speech related Applications....
From Huawei Canada - Wed, 17 Oct 2018 05:55:32 GMT - View all Montréal, QC jobs
          MACHINE LEARNING ENGINEER FOR SPEECH RELATED APPLICATIONS - Huawei Canada - Montréal, QC      Cache   Translate Page      
We thank all applicants for their interest in career opportunities with Huawei. Machine Learning Engineer for Speech related Applications (6 months contract)....
From Huawei Canada - Wed, 17 Oct 2018 05:55:32 GMT - View all Montréal, QC jobs
          Machine Learning Software Developer - Huawei Canada - Montréal, QC      Cache   Translate Page      
We thank all applicants for their interest in career opportunities with Huawei. ML Software developer....
From Huawei Canada - Fri, 05 Oct 2018 05:47:02 GMT - View all Montréal, QC jobs
          MACHINE LEARNING HARDWARE RESEARCHER OR DEVELOPER - Huawei Canada - Montréal, QC      Cache   Translate Page      
We thank all applicants for their interest in career opportunities with Huawei. Machine Learning Hardware Researcher or Developer....
From Huawei Canada - Fri, 05 Oct 2018 05:47:02 GMT - View all Montréal, QC jobs
          Introducing Storage Forecasting with SQL Server Machine Learning Services      Cache   Translate Page      

The newly released SentryOne 18.5 includes the first feature set to utilize advanced analytics and machine learning technology—introducing SentryOne Storage Forecasting. SentryOne Director of Analytics Steve Wright shares how you can access and customize storage capacity forecasts in SentryOne.

The post Introducing Storage Forecasting with SQL Server Machine Learning Services appeared first on SentryOne Team Blog.


          Using AI, machine learning in networking to improve analytics      Cache   Translate Page      
none
          ISV Technology Director - AI and ML - 67511 - Advanced Micro Devices, Inc. - Austin, TX      Cache   Translate Page      
AMD’s Machine Learning team work on many high-impact projects that serve AMD’s various lines of business. 3+ Years with leading ISV’s leveraging the Machine...
From Advanced Micro Devices, Inc. - Sun, 04 Nov 2018 07:32:49 GMT - View all Austin, TX jobs
          What Are Machine Learning Algorithms? Here’s How They Work      Cache   Translate Page      
machine-learning-algorithms

Artificial intelligence and machine learning produce many of the advancements we see in the technology industry today. But how are machines given the ability to learn? Furthermore, how does the way we do this result in unintended consequences? Here’s our quick explainer on how machine learning algorithms work, along with some examples of machine learning gone awry. What Are Machine Learning Algorithms? Machine learning is a branch of computer science that focuses on giving AI the ability to learn tasks. This includes developing abilities without programmers explicitly coding AI to do these things. Instead, the AI is able to use data to...

Read the full article: What Are Machine Learning Algorithms? Here’s How They Work


          Linux Systems Administrator - Kira Systems - Toronto, ON      Cache   Translate Page      
Growth potential, values and attitudes are equally important to be a successful Kiran. Kira Systems is a powerful machine learning software that identifies,...
From Indeed - Tue, 06 Nov 2018 16:04:32 GMT - View all Toronto, ON jobs
          Machine learning in healthcare: Software detects drug theft      Cache   Translate Page      
none
          Senior Technical Product Manager – Machine Learning - DefinedCrowd - Seattle, WA      Cache   Translate Page      
Familiarity with data storage technologies and patterns, distributed systems, data warehouse/data lake and ETL architectures with Hadoop ecosystem, such as...
From Definedcrowd - Tue, 06 Nov 2018 15:44:08 GMT - View all Seattle, WA jobs
          Marchex Acquires Telmetrics, For Up To $13.1M      Cache   Translate Page      

Marchex, Inc. (NASDAQ:MCHX), a Seattle, WA- and NYC-based provider of call analytics that drive, measure, and convert callers into customers, acquired Telmetrics, an enterprise call and text tracking and analytics company, for consideration of up to $13.1m in cash. By combining resources, the companies expect to leverage machine learning and AI-driven capabilities across a large […]

The post Marchex Acquires Telmetrics, For Up To $13.1M appeared first on FinSMEs.


          Machine Learning Engineer / Algorithm Developer - TECHNICA CORPORATION - Dulles, VA      Cache   Translate Page      
Job Description: We are seeking a highly creative software engineer experienced in artificial intelligence and deep learning techniques to design, develop,...
From Technica Corporation - Fri, 05 Oct 2018 10:31:19 GMT - View all Dulles, VA jobs
          Machine Learning Engineer - TECHNICA CORPORATION - Dulles, VA      Cache   Translate Page      
Technica Corporation is seeking a Machine Learning. Engineer to support our internal Innovation, Research....
From Technica Corporation - Thu, 23 Aug 2018 10:27:08 GMT - View all Dulles, VA jobs
          Open source machine learning tool could help choose cancer drugs      Cache   Translate Page      
(Georgia Institute of Technology) Using machine learning techniques, a new open source decision support tool could help clinicians choose cancer therapy drugs by analyzing RNA expression tied to information about patient outcomes with specific drugs.
          Data Engineer (Data Warehouse) - Bandwidth - Raleigh, NC      Cache   Translate Page      
Familiarity with Machine Learning &amp; Statistical concepts. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Mon, 05 Nov 2018 20:35:52 GMT - View all Raleigh, NC jobs
          Data Analyst, International - Bandwidth - Raleigh, NC      Cache   Translate Page      
Machine learning/AI techniques, features, and classifiers. Simply changing the way people communicate, connect and do business....
From Bandwidth - Thu, 04 Oct 2018 16:32:34 GMT - View all Raleigh, NC jobs
          Call for photos - we will buy new car photos      Cache   Translate Page      
We need a photos of following cars (all 2017-2018 generation): - Honda Civic - Ford Fiesta - Nissan Qashqai - Volkswagen Passat We need these photos for our internal project where we learn how to use Machine Learning in mobile apps; our aim is to build an app that will recognise cars... (Budget: $750 - $1500 USD, Jobs: Data Entry, Photography, Web Search)
          Audi mulls setting up Israel R&D center      Cache   Translate Page      
Globes: The German carmaker is building a fleet of vehicles in Europe equipped with software that includes machine learning, environmental perception and mapping skills
          Microsoft and Wal-Mart Partner to Take on Amazon      Cache   Translate Page      
Quite interesting details in the complete article linked to below.   Microsoft has the cloud based technical capabilities.  Also the IOT architecture that will be important.    But is the advanced tech that Amazon has installed already well ahead?  The partnership makes sense to test that.

Microsoft and Wal-mart are creating a ‘cloud factory’ to take on Amazon   By Mike Wheatley in SiliconAngle

Microsoft Corp. and Wal-Mart Stores Inc. are building on a strategic partnership announced in July that saw them commit to using the Redmond software giant’s cloud, artificial intelligence and “internet of things” tools to modernize the retailer’s business operations.

Microsoft and Walmart today said they’ve created a new “cloud factory” at the latter’s existing Innovation Hub (pictured) in Austin, Texas. Known as “4.co” due to its location on the corner of Fourth and Colorado streets, the joint engineering facility is set to open early next year and will be staffed by a team of 30 technologists from both companies.

One of the goals at the facility will be to help Walmart move thousands of its internal business applications over to Microsoft’s Azure cloud platform. The engineers will also work together to develop brand-new, cloud-native applications. In order to do so, the companies will make use of Microsoft’s cognitive services, chatbot and machine learning tools, Clay Johnson, executive vice president and enterprise chief information officer at Walmart, said in an interview with Microsoft Transform.  .... "


          Model Extraction and Active Learning. (arXiv:1811.02054v1 [cs.LG])      Cache   Translate Page      

Authors: Varun Chandrasekaran, Kamalika Chaudhuri, Irene Giacomelli, Somesh Jha, Songbai Yan

Machine learning is being increasingly used by individuals, research institutions, and corporations. This has resulted in the surge of Machine Learning-as-a-Service (MLaaS) - cloud services that provide (a) tools and resources to learn the model, and (b) a user-friendly query interface to access the model. However, such MLaaS systems raise privacy concerns, one being model extraction. Adversaries maliciously exploit the query interface to steal the model. More precisely, in a model extraction attack, a good approximation of a sensitive or proprietary model held by the server is extracted (i.e. learned) by a dishonest user. Such a user only sees the answers to select queries sent using the query interface. This attack was recently introduced by Tramer et al. at the 2016 USENIX Security Symposium, where practical attacks for different models were shown. We believe that better understanding the efficacy of model extraction attacks is paramount in designing better privacy-preserving MLaaS systems. To that end, we take the first step by (a) formalizing model extraction and proposing the first definition of extraction defense, and (b) drawing parallels between model extraction and the better investigated active learning framework. In particular, we show that recent advancements in the active learning domain can be used to implement both model extraction, and defenses against such attacks.


          "I had a solid theory before but it's falling apart": Polarizing Effects of Algorithmic Transparency. (arXiv:1811.02163v1 [cs.HC])      Cache   Translate Page      

Authors: Aaron Springer, Steve Whittaker

The rise of machine learning has brought closer scrutiny to intelligent systems, leading to calls for greater transparency and explainable algorithms. We explore the effects of transparency on user perceptions of a working intelligent system for emotion detection. In exploratory Study 1, we observed paradoxical effects of transparency which improves perceptions of system accuracy for some participants while reducing accuracy perceptions for others. In Study 2, we test this observation using mixed methods, showing that the apparent transparency paradox can be explained by a mismatch between participant expectations and system predictions. We qualitatively examine this process, indicating that transparency can undermine user confidence by causing users to fixate on flaws when they already have a model of system operation. In contrast transparency helps if users lack such a model. Finally, we revisit the notion of transparency and suggest design considerations for building safe and successful machine learning systems based on our insights.


          Hybrid Approach to Automation, RPA and Machine Learning: a Method for the Human-centered Design of Software Robots. (arXiv:1811.02213v1 [cs.SE])      Cache   Translate Page      

Authors: Wiesław Kopeć, Marcin Skibiński, Cezary Biele, Kinga Skorupska, Dominika Tkaczyk, Anna Jaskulska, Katarzyna Abramczuk, Piotr Gago, Krzysztof Marasek

One of the more prominent trends within Industry 4.0 is the drive to employ Robotic Process Automation (RPA), especially as one of the elements of the Lean approach. The full implementation of RPA is riddled with challenges relating both to the reality of everyday business operations, from SMEs to SSCs and beyond, and the social effects of the changing job market. To successfully address these points there is a need to develop a solution that would adjust to the existing business operations and at the same time lower the negative social impact of the automation process.

To achieve these goals we propose a hybrid, human-centered approach to the development of software robots. This design and implementation method combines the Living Lab approach with empowerment through participatory design to kick-start the co-development and co-maintenance of hybrid software robots which, supported by variety of AI methods and tools, including interactive and collaborative ML in the cloud, transform menial job posts into higher-skilled positions, allowing former employees to stay on as robot co-designers and maintainers, i.e. as co-programmers who supervise the machine learning processes with the use of tailored high-level RPA Domain Specific Languages (DSLs) to adjust the functioning of the robots and maintain operational flexibility.


          Solutions Architect - NVIDIA - Washington State      Cache   Translate Page      
Assist field business development in through the enablement process for GPU Computing products, technical relationship and assisting machine learning/deep...
From NVIDIA - Sun, 19 Aug 2018 07:55:46 GMT - View all Washington State jobs
          Solutions Architect - NVIDIA - Virginia      Cache   Translate Page      
Build and cultivate internal understanding of data analytics and machine learning among the NVIDIA technical community....
From NVIDIA - Tue, 02 Oct 2018 07:57:56 GMT - View all Virginia jobs
          US Federal CTO - Solution Architect - NVIDIA - Virginia      Cache   Translate Page      
Build and cultivate internal understanding of Federal customer data analytics and machine learning requirements among the NVIDIA technical community including...
From NVIDIA - Tue, 07 Aug 2018 01:54:32 GMT - View all Virginia jobs
          Fullstack Software Engineer | New York - BenevolentAI - New York, NY      Cache   Translate Page      
Machine Learning Squads. All employment is decided on the basis of qualifications, merit, and business need. Fun internal events (boat parties, karting, Oktober...
From BenevolentAI - Sat, 27 Oct 2018 10:26:13 GMT - View all New York, NY jobs
          Solutions Architect - NVIDIA - Maryland      Cache   Translate Page      
You will also be an internal champion for Data Analytics and Machine Learning among the NVIDIA technical community....
From NVIDIA - Tue, 06 Nov 2018 07:57:54 GMT - View all Maryland jobs
          Sr. Solutions Architect - Team Lead - NVIDIA - Maryland      Cache   Translate Page      
You will also be an internal champion for Data Analytics and Machine Learning among the NVIDIA technical community....
From NVIDIA - Wed, 12 Sep 2018 19:54:41 GMT - View all Maryland jobs
          Senior Solution Architect - Cyber Security - NVIDIA - Maryland      Cache   Translate Page      
You will also be an internal champion for Cyber Security and Machine Learning among the NVIDIA technical community....
From NVIDIA - Wed, 01 Aug 2018 07:57:49 GMT - View all Maryland jobs
          Sr. Solutions Architect - Lead - NVIDIA - Washington, DC      Cache   Translate Page      
You will also be an internal champion for Data Analytics and Machine Learning among the NVIDIA technical community....
From NVIDIA - Wed, 12 Sep 2018 19:54:40 GMT - View all Washington, DC jobs
          High Dimensional Clustering with $r$-nets. (arXiv:1811.02288v1 [cs.CG])      Cache   Translate Page      

Authors: Georgia Avarikioti, Alain Ryser, Yuyi Wang, Roger Wattenhofer

Clustering, a fundamental task in data science and machine learning, groups a set of objects in such a way that objects in the same cluster are closer to each other than to those in other clusters. In this paper, we consider a well-known structure, so-called $r$-nets, which rigorously captures the properties of clustering. We devise algorithms that improve the run-time of approximating $r$-nets in high-dimensional spaces with $\ell_1$ and $\ell_2$ metrics from $\tilde{O}(dn^{2-\Theta(\sqrt{\epsilon})})$ to $\tilde{O}(dn + n^{2-\alpha})$, where $\alpha = \Omega({\epsilon^{1/3}}/{\log(1/\epsilon)})$. These algorithms are also used to improve a framework that provides approximate solutions to other high dimensional distance problems. Using this framework, several important related problems can also be solved efficiently, e.g., $(1+\epsilon)$-approximate $k$th-nearest neighbor distance, $(4+\epsilon)$-approximate Min-Max clustering, $(4+\epsilon)$-approximate $k$-center clustering. In addition, we build an algorithm that $(1+\epsilon)$-approximates greedy permutations in time $\tilde{O}((dn + n^{2-\alpha}) \cdot \log{\Phi})$ where $\Phi$ is the spread of the input. This algorithm is used to $(2+\epsilon)$-approximate $k$-center with the same time complexity.


          Sr Data Scientist - NVIDIA - Santa Clara, CA      Cache   Translate Page      
3+ years’ experience in solving problems using machine learning algorithms and techniques (clustering, classification, outlier analysis, etc.)....
From NVIDIA - Tue, 30 Oct 2018 01:54:49 GMT - View all Santa Clara, CA jobs
          Kalman Filter Modifier for Neural Networks in Non-stationary Environments. (arXiv:1811.02361v1 [cs.LG])      Cache   Translate Page      

Authors: Honglin Li, Frieder Ganz, Shirin Enshaeifar, Payam Barnaghi

Learning in a non-stationary environment is an inevitable problem when applying machine learning algorithm to real world environment. Learning new tasks without forgetting the previous knowledge is a challenge issue in machine learning. We propose a Kalman Filter based modifier to maintain the performance of Neural Network models under non-stationary environments. The result shows that our proposed model can preserve the key information and adapts better to the changes. The accuracy of proposed model decreases by 0.4% in our experiments, while the accuracy of conventional model decreases by 90% in the drifts environment.


          Artificial Intelligence / Machine Learning Experts - Fujitsu Technology Solutions - Anderlecht      Cache   Translate Page      
Looking for Artificial Intelligence / Machine Learning Experts - Data Scientists, Consultants, Architects, Developers and Business Analysts Contribute some (Artificial) Intelligence? - Join Fujitsu's digital business team! Artificial Intelligence is the New Black, our customers are feeling the breeze of change, and have request our Intelligence to their assistance. We are looking to recruit several ready professionals and young talents to be the experts and help our customers to...
          Mobile Data Science: Towards Understanding Data-Driven Intelligent Mobile Applications. (arXiv:1811.02491v1 [cs.CY])      Cache   Translate Page      

Authors: Iqbal H. Sarker

Due to the popularity of smart mobile phones and context-aware technology, various contextual data relevant to users' diverse activities with mobile phones is available around us. This enables the study on mobile phone data and context-awareness in computing, for the purpose of building data-driven intelligent mobile applications, not only on a single device but also in a distributed environment for the benefit of end users. Based on the availability of mobile phone data, and the usefulness of data-driven applications, in this paper, we discuss about mobile data science that involves in collecting the mobile phone data from various sources and building data-driven models using machine learning techniques, in order to make dynamic decisions intelligently in various day-to-day situations of the users. For this, we first discuss the fundamental concepts and the potentiality of mobile data science to build intelligent applications. We also highlight the key elements and explain various key modules involving in the process of mobile data science. This article is the first in the field to draw a big picture, and thinking about mobile data science, and it's potentiality in developing various data-driven intelligent mobile applications. We believe this study will help both the researchers and application developers for building smart data-driven mobile applications, to assist the end mobile phone users in their daily activities.


          Interpretation of Neural Networks is Fragile. (arXiv:1710.10547v2 [stat.ML] UPDATED)      Cache   Translate Page      

Authors: Amirata Ghorbani, Abubakar Abid, James Zou

In order for machine learning to be deployed and trusted in many applications, it is crucial to be able to reliably explain why the machine learning algorithm makes certain predictions. For example, if an algorithm classifies a given pathology image to be a malignant tumor, then the doctor may need to know which parts of the image led the algorithm to this classification. How to interpret black-box predictors is thus an important and active area of research. A fundamental question is: how much can we trust the interpretation itself? In this paper, we show that interpretation of deep learning predictions is extremely fragile in the following sense: two perceptively indistinguishable inputs with the same predicted label can be assigned very different interpretations. We systematically characterize the fragility of several widely-used feature-importance interpretation methods (saliency maps, relevance propagation, and DeepLIFT) on ImageNet and CIFAR-10. Our experiments show that even small random perturbation can change the feature importance and new systematic perturbations can lead to dramatically different interpretations without changing the label. We extend these results to show that interpretations based on exemplars (e.g. influence functions) are similarly fragile. Our analysis of the geometry of the Hessian matrix gives insight on why fragility could be a fundamental challenge to the current interpretation approaches.


          Machine Learning for Set-Identified Linear Models. (arXiv:1712.10024v2 [stat.ML] UPDATED)      Cache   Translate Page      

Authors: Vira Semenova

Set-identified models often restrict the number of covariates leading to wide identified sets in practice. This paper provides estimation and inference methods for set-identified linear models with high-dimensional covariates where the model selection is based on modern machine learning tools. I characterize the boundary (i.e, support function) of the identified set using a semiparametric moment condition. Combining Neyman-orthogonality and sample splitting ideas, I construct a root-N consistent, the uniformly asymptotically Gaussian estimator of the support function. I also prove the validity of the Bayesian bootstrap procedure to conduct inference about the identified set. I provide a general method to construct a Neyman-orthogonal moment condition for the support function. I apply this result to estimate sharp nonparametric bounds on the average treatment effect in Lee (2008)'s model of endogenous selection and substantially tighten the bounds on this parameter in Angrist et al. (2006)'s empirical setting. I also apply this result to estimate sharp identified sets for two other parameters - a new parameter, called a partially linear predictor, and the average partial derivative when the outcome variable is recorded in intervals.


          Artificial Intelligence Enabled Software Defined Networking: A Comprehensive Overview. (arXiv:1803.06818v3 [cs.AI] UPDATED)      Cache   Translate Page      

Authors: Majd Latah, Levent Toker

Software defined networking (SDN) represents a promising networking architecture that combines central management and network programmability. SDN separates the control plane from the data plane and moves the network management to a central point, called the controller, that can be programmed and used as the brain of the network. Recently, the research community has showed an increased tendency to benefit from the recent advancements in the artificial intelligence (AI) field to provide learning abilities and better decision making in SDN. In this study, we provide a detailed overview of the recent efforts to include AI in SDN. Our study showed that the research efforts focused on three main sub-fields of AI namely: machine learning, meta-heuristics and fuzzy inference systems. Accordingly, in this work we investigate their different application areas and potential use, as well as the improvements achieved by including AI-based techniques in the SDN paradigm.


          (USA-CA-San Jose) Sr Data Architect – Big Data      Cache   Translate Page      
**Danaher Company Description** Danaher is a global science & technology innovator committed to helping our customers solve complex challenges and improve quality of life worldwide. Our world class brands are leaders in some of the most demanding and attractive industries, including life sciences, medical diagnostics, dental, environmental and applied solutions. Our globally diverse team of 67,000 associates is united by a common culture and operating system, the Danaher Business System, which serves as our competitive advantage. We generated $18.3B in revenue last year. We are ranked #162 on the Fortune 500 and our stock has outperformed the S&P 500 by more than 1,200% over 20 years. At Danaher, you can build a career in a way no other company can duplicate. Our brands allow us to offer dynamic careers across multiple industries. We’re innovative, fast-paced, results-oriented, and we win. We need talented people to keep winning. Here you’ll learn how DBS is used to shape strategy, focus execution, align our people, and create value for customers and shareholders. Come join our winning team. **Description** *Danaher Digital* Danaher Digital is our digital innovation and acceleration center where we’re bringing together the leading strategic product and business leaders, technologists and data scientists for the common purpose of accelerating development and commercialization of disruptive and transformative digital solutions into the marketplace. We accelerate Danaher’s digital innovation journey by partnering with Danaher operating companies (OpCos) to monetize and commercialize the potential of emerging digital trends. Located in Silicon Valley, the heart of global innovation, Danaher Digital is ideally situated to capitalize on the digital mega trends transforming our world, including Internet-of Things (IoT), Data, AI, cloud, mobile, Augmented Reality (AR), Blockchain and other Digital frontiers. *Senior Data Architect* You will report to the Senior Director of Data & Analytics and will be responsible for leading the vision, design, development and deployment of large-scale data fabrics and data platforms for Danaher’s IoT and Analytics Machine Learning solutions. The right candidate will provide strategic and technical leadership in using best-of-breed big data technologies with the objective of bringing to market diverse solutions in IoT and Analytics for health sciences, medical diagnostics, industrial and other markets. This person will use his/her Agile experience to work collaboratively with other Product Managers/Owners in geographically distributed teams. *Responsibilities*: * Lead analysis, architecture, design and development of large-scale data fabrics and data platforms Advanced Analytics and IoT solutions based on best-of-breed and contemporary big-data technologies * Provide strategic leadership in evaluation, selection and/or architecture of modern data stacks supporting diverse solutions including Advanced analytics for health sciences, medical diagnostics, industrial and other markets * Provide technical leadership and delivery in all phases of a solution design from a data perspective: discovery, planning, implementation and data operations * Manage the full life-cycle of data - ingestion, aggregation, storage, access and security - for IoT and advanced analytics solutions * Work collaboratively with Product Management and Product Owners from other business units and/or customers to translate business requirements in to technical requirements (Epics and stories) in Agile process to drive data architecture * Own and drive contemporary data architecture and technology vision/road map while keeping abreast of the technology advances and architectural best practices **Qualification** *Required Skills & Experience: * * Bachelor’s degree in Computer Science or related field of study * Experience with security configurations of the Data Stack * 7 years’ hands-on leader in designing, building and running successful full stack Big Data, Data Fabric and/or Platforms and IoT/Analytics solutions in production environments * Must have experience in major big data solutions like Hadoop, HBase, Sqoop, Hive, Spark etc. * Architected, developed and deployed data solutions on one or more of: AWS, Azure or Google IaaS/PaaS services for data, analytics and visualization * Demonstrated experience of full IoT and Advanced Analytics data technology stack, including data capture, ingestion, storage, analytics and visualization * Has worked with customers as a trusted advisor in Data architecture and management. * Working experience with ETL tools, storage infrastructure, streaming and batch data, data quality tools, data modeling tools, data integration and data visualization tools * Must have experience in leading projects in Agile development methodologies * Provide mentorship and thought leadership for immediate and external(customer) teams in best practices of Data platform; Lead conversations around extracting value out of Data Platform * Travel up to 40% required **Danaher Corporation Overview** Danaher is a global science & technology innovator committed to helping our customers solve complex challenges and improve quality of life worldwide. Our world class brands are leaders in some of the most demanding and attractive industries, including life sciences, medical diagnostics, dental, environmental and applied solutions. Our globally diverse team of 67,000 associates is united by a common culture and operating system, the Danaher Business System, which serves as our competitive advantage. We generated $18.3B in revenue last year. We are ranked #162 on the Fortune 500 and our stock has outperformed the S&P 500 by more than 1,200% over 20 years. At Danaher, you can build a career in a way no other company can duplicate. Our brands allow us to offer dynamic careers across multiple industries. We’re innovative, fast-paced, results-oriented, and we win. We need talented people to keep winning. Here you’ll learn how DBS is used to shape strategy, focus execution, align our people, and create value for customers and shareholders. Come join our winning team. **Organization:** Corporate **Job Function:** Information Technology **Primary Location:** North America-North America-United States-CA-San Jose **Schedule:** Full-time **Req ID:** COR001259
          geomstats: a Python Package for Riemannian Geometry in Machine Learning. (arXiv:1805.08308v2 [cs.LG] UPDATED)      Cache   Translate Page      

Authors: Nina Miolane, Johan Mathe, Claire Donnat, Mikael Jorda, Xavier Pennec

We introduce geomstats, a python package that performs computations on manifolds such as hyperspheres, hyperbolic spaces, spaces of symmetric positive definite matrices and Lie groups of transformations. We provide efficient and extensively unit-tested implementations of these manifolds, together with useful Riemannian metrics and associated Exponential and Logarithm maps. The corresponding geodesic distances provide a range of intuitive choices of Machine Learning loss functions. We also give the corresponding Riemannian gradients. The operations implemented in geomstats are available with different computing backends such as numpy, tensorflow and keras. We have enabled GPU implementation and integrated geomstats manifold computations into keras deep learning framework. This paper also presents a review of manifolds in machine learning and an overview of the geomstats package with examples demonstrating its use for efficient and user-friendly Riemannian geometry.


          ENG: End-to-end Neural Geometry for Robust Depth and Pose Estimation using CNNs. (arXiv:1807.05705v2 [cs.CV] UPDATED)      Cache   Translate Page      

Authors: Thanuja Dharmasiri, Andrew Spek, Tom Drummond

Recovering structure and motion parameters given a image pair or a sequence of images is a well studied problem in computer vision. This is often achieved by employing Structure from Motion (SfM) or Simultaneous Localization and Mapping (SLAM) algorithms based on the real-time requirements. Recently, with the advent of Convolutional Neural Networks (CNNs) researchers have explored the possibility of using machine learning techniques to reconstruct the 3D structure of a scene and jointly predict the camera pose. In this work, we present a framework that achieves state-of-the-art performance on single image depth prediction for both indoor and outdoor scenes. The depth prediction system is then extended to predict optical flow and ultimately the camera pose and trained end-to-end. Our motion estimation framework outperforms the previous motion prediction systems and we also demonstrate that the state-of-the-art metric depths can be further improved using the knowledge of pose.


          Reinforcement Learning: An Introduction (adaptive Computation And Machine Learning) Second Edition Edition      Cache   Translate Page      
Richard S. Sutton / Science / 2018
          The Curse of Concentration in Robust Learning: Evasion and Poisoning Attacks from Concentration of Measure. (arXiv:1809.03063v2 [cs.LG] UPDATED)      Cache   Translate Page      

Authors: Saeed Mahloujifar, Dimitrios I. Diochnos, Mohammad Mahmoody

Many modern machine learning classifiers are shown to be vulnerable to adversarial perturbations of the instances. Despite a massive amount of work focusing on making classifiers robust, the task seems quite challenging. In this work, through a theoretical study, we investigate the adversarial risk and robustness of classifiers and draw a connection to the well-known phenomenon of concentration of measure in metric measure spaces. We show that if the metric probability space of the test instance is concentrated, any classifier with some initial constant error is inherently vulnerable to adversarial perturbations.

One class of concentrated metric probability spaces are the so-called Levy families that include many natural distributions. In this special case, our attacks only need to perturb the test instance by at most $O(\sqrt n)$ to make it misclassified, where $n$ is the data dimension. Using our general result about Levy instance spaces, we first recover as special case some of the previously proved results about the existence of adversarial examples. However, many more Levy families are known (e.g., product distribution under the Hamming distance) for which we immediately obtain new attacks that find adversarial examples of distance $O(\sqrt n)$.

Finally, we show that concentration of measure for product spaces implies the existence of forms of "poisoning" attacks in which the adversary tampers with the training data with the goal of degrading the classifier. In particular, we show that for any learning algorithm that uses $m$ training examples, there is an adversary who can increase the probability of any "bad property" (e.g., failing on a particular test instance) that initially happens with non-negligible probability to $\approx 1$ by substituting only $\tilde{O}(\sqrt m)$ of the examples with other (still correctly labeled) examples.


          Optimal Weighting for Exam Composition. (arXiv:1801.06043v1 [cs.CY] CROSS LISTED)      Cache   Translate Page      

Authors: Sam Ganzfried, Farzana Yusuf

A problem faced by many instructors is that of designing exams that accurately assess the abilities of the students. Typically these exams are prepared several days in advance, and generic question scores are used based on rough approximation of the question difficulty and length. For example, for a recent class taught by the author, there were 30 multiple choice questions worth 3 points, 15 true/false with explanation questions worth 4 points, and 5 analytical exercises worth 10 points. We describe a novel framework where algorithms from machine learning are used to modify the exam question weights in order to optimize the exam scores, using the overall class grade as a proxy for a student's true ability. We show that significant error reduction can be obtained by our approach over standard weighting schemes, and we make several new observations regarding the properties of the "good" and "bad" exam questions that can have impact on the design of improved future evaluation methods.


          Support system for ATLAS distributed computing operations      Cache   Translate Page      
The ATLAS distributed computing system has allowed the experiment to successfully meet the challenges of LHC Run 2. In order for distributed computing to operate smoothly and efficiently, several support teams are organized in the ATLAS experiment. The ADCoS is a dedicated group of shifters who follow and report failing jobs, failing data transfers between sites, degradation of ATLAS central computing services, and more. The DAST provides user support to resolve issues related to running distributed analysis on the Grid. The CRC maintains a global view of the day- to-day operations. In this paper, the status and operational experience of the support system for ATLAS distributed computing in LHC Run 2 are reported. This report also includes operation experience from the Grid site point of view, and an analysis of the errors that create the biggest waste of wallclock time. The report of operation experience will focus on some of the more time-consuming tasks for shifters, and on the introduction of new technologies, such as machine learning, to ease the work.
          Machine Learning Researcher - PARC, a Xerox company - Palo Alto, CA      Cache   Translate Page      
PARC, a Xerox company, is in the Business of Breakthroughs®. We create new business options, accelerate time to market, augment internal capabilities, and...
From PARC, a Xerox company - Sun, 26 Aug 2018 12:12:32 GMT - View all Palo Alto, CA jobs
          Interactive Machine Learning Researcher - PARC, a Xerox company - Palo Alto, CA      Cache   Translate Page      
PARC, a Xerox company, is in the Business of Breakthroughs®. We create new business options, accelerate time to market, augment internal capabilities, and...
From PARC, a Xerox company - Sun, 26 Aug 2018 12:12:32 GMT - View all Palo Alto, CA jobs
          Comment on Top 10 Artificial Intelligence (AI) Technologies In 2019 by Impact of Artificial Intelligence (AI) In Medicine and Biology      Cache   Translate Page      
[…] The broad uses of Artificial Intelligence (AI) Technologies applications have been already been noticed by all of us in several areas of developments with main impacts on technology-based stuff including machine learning devices, robotics etc to name a few. Artificial Intelligence (AI) is one of the most famously used cutting-edge technologies whose importance major to exhibit human intelligence kind of behavior has been noticed and accordingly almost all the fields have been seen to implement the uses of artificial intelligence based systems. The technology makes any operating device equipped enough to expose human-like behavior with self-understanding and decision-making capabilities and accordingly putting the stuff in front of us. Following the wide versatility of this cutting-edge technology extremely famous with the name Artificial Intelligence, health care based sectors associated with the fields of medicine and biology have also attempted to implement the use of artificial intelligence in the developmental procedures. The implementations and benefits of Artificial Intelligence (AI) in medicine and biology can be understood with the help of following bullet points. Have you gone through these top Artificial Intelligence (AI) Technologies? […]
          Imaging and Machine Learning Research Scientist - AIRY:3D - Montréal, QC      Cache   Translate Page      
Contribute to expanding our patent portfolio, generate publishable research. We’re looking to add even more imaging and machine learning expertise to our team....
From Indeed - Wed, 17 Oct 2018 15:54:14 GMT - View all Montréal, QC jobs
          BXB Digital: IoT Technical Program Manager      Cache   Translate Page      
CA-Santa Clara, Job ID #: 7949 BXB Digital connects product conveyance platforms with digital capabilities to create more connected, intelligent and efficient supply chains. Established in 2016, BXB Digital is the newest business unit within the Brambles family, the global leader in supply chain logistics. BXB Digital combines network concepts, enterprise supply chain expertise, machine learning and the Internet
          The rise of local mapping communities      Cache   Translate Page      
Members of the mapping community in Kinshasa plan the collection of field data for the Kisenso neighborhood. (courtesy of OpenDRI)
Members of the mapping community in Kinshasa, DR Congo plan the collection of field data for the Kisenso neighborhood. (Courtesy of OpenDRI)

There is a unique space where you can encounter everyone from developers of self-driving cars in Silicon Valley to city planners in Niamey to humanitarian workers in Kathmandu Valley: the global OpenStreetMap (OSM) community. It comprises a geographically and experientially diverse network of people who contribute to OSM, a free and editable map of the world that is often called the “Wikipedia of maps.”  

What is perhaps most special about this community is its level playing field. Anyone passionate about collaborative mapping can have a voice from anywhere in the world. In the past few years, there has been a meteoric rise of locally organized mapping communities in developing countries working to improve the map in service of sustainable development activities.

The next opportunity to see the OSM community in action will be the November 14th mapathon hosted by the Global Facility for Disaster Reduction and Recovery (GFDRR)’s Open Data for Resilience Initiative (OpenDRI). Mapathons bring together volunteers to improve the maps of some of the world’s most vulnerable areas, not only easing the way for emergency responders when disaster strikes, but also helping cities and communities plan and build more resiliently for the future.

GFDRR’s engagement with local OSM communities

[[tweetable]]The 2010 Haiti earthquake served as a wake-up call about the need for access to better quality information for reducing vulnerability to natural hazards and climate change impacts.[[/tweetable]] In the years since, OpenDRI has turned to the OSM platform as an important way to bring people together to create open data, learn new skills, and support the human networks that eventually become key actors for resilience. We can gather people in a room around something exciting, like a mapathon, and start a conversation about sharing information for the benefit of everyone.

Changes in the mapped areas in OpenStreetMap for Kampala, Uganda, from 2016 to 2018. (courtesy of OpenDRI and OSM)
Changes in the mapped areas in OpenStreetMap for Kampala, Uganda, from 2016 to 2018. (Courtesy of OpenDRI and OSM)
[[tweetable]]Any data, technology, or tool is only as valuable as the way and the extent to which people use it[[/tweetable]], and that’s why building sustainable mapping communities is so critical for this work. Even as we engage governments to promote the use of open data and open source tools, OpenDRI also strives to nurture local communities of OSM users and developers from universities, NGOs, and innovation hubs. To that end, OpenDRI supports local OSM communities and conferences like “State of the Map” whenever possible, particularly by funding scholarships for attendees who would not otherwise get to attend, learn, and share knowledge.

Participatory mapping in Asia and Africa
A member of the local mapping community in Uganda collects field data for OSM. (Courtesy of OpenDRI)
A member of the local mapping community in Uganda
collects field data for OSM. (Courtesy of OpenDRI)


OpenDRI started its work with OSM by supporting the growth of local mapping communities in  Indonesia, the Philippines, Nepal, Bangladesh, and Sri Lanka, including through the Open Cities Project. Many of these communities were quick on their feet to respond to the devastating 2015 earthquakes in Nepal. More than 6,000 volunteers helped add data to the OSM platform, mapping up to 80 percent of affected zones, an effort which continues to provide invaluable information to emergency response and preparedness efforts.In the years since, OSM communities across Asia have come together to exchange knowledge and build connections at a series of open source mapping conferences. The fourth “State of the Map Asia” conference will take place in Bangalore, India, this month.

In Africa, the stakes for OSM are even higher, because it is often the only digital map available for many locations. Recent years have brought a rise of participatory mapping communities across Africa, which now total more than 30 active local OSM groups. Africa’s first-ever “State of the Map” conference was held in Kampala, Uganda in 2017.

Building on that momentum, OpenDRI recently launched the Open Cities Africa project, currently supporting the development of teams in 11 cities across Africa. These teams are taking the lead in collecting data remotely and on-the-ground through participatory mapping, thus building mapping capacity in their local OSM communities. They are also collaborating with World Bank teams to use the new OSM data to help address a range of development challenges, from urban flooding in Kinshasa to coastal risk management in Senegal. Drawing on our experiences in Asia, we are incorporating novel approaches in our engagement in Africa, including online learning, gender integration, disruptive technologies, and design research.

What’s next for local OSM communities?

About this series
More blog posts


Local OSM communities are hopeful that the future will see a larger and more diverse population of mappers worldwide – this will be key to improving the “Wikipedia of maps” even further. As technology giants join the global OSM community, we are now exploring how new machine learning mapping techniques might complement and amplify the work of local OSM communities.

Over the past seven years, the OpenDRI team has been hard at work to create local communities around open-source mapping as part of our drive to promote open data for resilience, and that effort will continue.

To discover the OSM community for yourself and learn more about the benefits of using geospatial data for addressing the world’s most critical development challenges, join us on Wednesday, November 14 for the OpenDRI mapathon at the World Bank.  
 
READ MORE
          Director, Marketing Automation for Relationship Marketing - Microsoft - Redmond, WA      Cache   Translate Page      
In addition, this role will lead the identification and leveraging of the latest marketing capabilities like machine learning and AI to propel Relationship...
From Microsoft - Wed, 31 Oct 2018 07:57:36 GMT - View all Redmond, WA jobs
          German anti-fraud startup Fraugster raises $14 million      Cache   Translate Page      

Berlin-based startup Fraugster that applies AI and machine learning algorithms to fraud detection in the e-commerce space, has raised $14 million in a funding round led by CommerzVentures, the venture capital subsidiary of Commerzbank, and early Fraugster investors Earlybird, Speedinvest, Seedcamp and Rancilio Cube, with participation from HSB Ventures, the venture capital arm of the […]

The post German anti-fraud startup Fraugster raises $14 million appeared first on Tech.eu.


          The Koch Brothers Are Watching You -- And New Documents Reveal Just How Much They Know      Cache   Translate Page      
Billionaire brothers have built personality profiles of most Americans, and use them to push right-wing propaganda

New documents uncovered by the Center for Media and Democracy show that the billionaire Koch brothers have developed detailed personality profiles on 89 percent of the U.S. population; and are using those profiles to launch an unprecedented private propaganda offensive to advance Republican candidates in the 2018 midterms.

The documents also show that the Kochs have developed persuasion models — like their "Heroin Model" and "Heroin Treatment Model" — that target voters with tailored messaging on select issues, and partner with cable and satellite TV providers to play those tailored messages during “regular” television broadcasts.

Over the last decade, big data and microtargeting have revolutionized political communications. And the Kochs, who are collectively worth $120 billion, now stand at the forefront of that revolution — investing billions in data aggregation, machine learning, software engineering and Artificial Intelligence optimization.

In modern elections, incorporating AI into voter file maintenance has become a prerequisite to producing reliable data. The Kochs’ political data firm, i360 states that it has “been practicing AI for years. Our team of data scientists uses components of Machine learning, Deep Learning and Predictive Analytics, every day as they build and refine our predictive models.”

Thanks to that investment (and the Supreme Court’s campaign finance rulings that opened the floodgates for super PACs), the Koch network is better positioned than either the Democratic Party or the GOP to reach voters with their individually tailored communications.

That is a dangerous development, with potentially dramatic consequences for our democracy.

The Kochs and i360

The Kochs formally entered the data space nine years ago, developing the “Themis Trust” program for the 2010 midterms — an uncommonly impactful election cycle where Republican operatives executed their REDMAP program and algorithmically gerrymandered congressional maps across the country in their favor.

In 2011, the Kochs folded Themis into a data competitor it acquired, i360 LLC, which was founded by Michael Palmer, the former chief technology officer of Sen. John McCain’s 2008 presidential campaign. Palmer still leads the organization.

Back then, as journalists Kenneth Vogel and Mike Allen documented, the Kochs’ long-term funding commitments to i360 allowed the organization to think bigger than their political competitors.

“Right now, we’re talking about and building things that you won’t see in 2016, because it’s not going to be ready until 2018,” Michael Palmer said in the wake of the 2014 midterm cycle.

Those programs are now operational. And according to a successful GOP campaign manager, i360 is the “best in the business” at providing Republicans with voter data.

i360’s client list reflects that data superiority. The country’s most notorious and effective political spenders, like the National Rifle Association, use the platform to identify and influence voters, as do Republican party committees, and U.S. House and Senate campaigns.

(A full list of i360’s clients is available here. Some clients, like the Republican Party of Wisconsin, have multiple sub-campaigns they run. It is also important to note that many Koch political groups, like Americans for Prosperity and the Libre Initiative, signed data sharing agreements with i360 in 2016 that are most likely still in effect.)

i360 sweetens the deal to its clients by offering its services at below-market rates. And once clients are locked into the i360 platform, they have access to the company’s voter file — the beating heart of modern political campaigns.

Conservatives agree that the Kochs are subsidizing i360. The losses they sustain by undercharging clients, however, are a pittance compared to the down-stream public policy returns and political power the Kochs receive from operating what amounts to a shadow political party in the United States — one that vigilantly guards the fossil fuel subsidies, deregulatory schemes, and regressive tax structures that enable Koch Industries to bring in $115 billion annually in private revenue.

Inside the i360 Voter File

i360’s voter file identifies “more than 199 million active voters and 290 million U.S. consumers,” and provides its users with up to 1,800 unique data points on each identified individual.

As a result, i360 and the Kochs know your vitals, ethnicity, religion, occupation, hobbies, shopping habits, political leanings, financial assets, marital status and much more.

They know if you enjoy fishing — and if you do, whether you prefer salt or fresh water. They know if you have bladder control difficulty, get migraines or have osteoporosis. They know which advertising mediums (radio, TV, internet, email) are the most effective. For you.

i360 has the following attribute tags, among hundreds of others, ranked 1-10, or subdivided otherwise in their voter file.

Here’s an example of an i360 attribute tag and code name, using a 1-10 value scale:

But i360 attribute codes are not limited to that 1-10 scale. Their knowledge of your financial standing is granular, from how much equity you have in your home to your net wealth and expendable income.

They know where you live, what your mortgage status is and even how many bathrooms are in your house.

i360 has also created a set of 70 “clustercodes” to humanize its data for campaign operatives. These categories range from “Faded Blue Collars” to “Meandering Millennials,” and have flamboyant descriptions that correspond with their attribute headings.

Here are