Next Page: 10000

          Roasted corn — and machine learning in a food truck — at 23rd and Union’s El Costeño      Cache   Translate Page      
Moises Santos is a 24-year-old programmer, food truck designer, and immigrant from Oaxaca, Mexico. His food truck holds down what seems like prime territory — the pot purchasing and stoner friendly parking lot at the Central District’s Uncle Ike’s. The … Continue reading
          03-25-2019 CS Colloquium: Anand Iyer (University of California, Berkeley) - Scalable Systems for Large-Scale Dynamic Connected Data Processing       Cache   Translate Page      
Speaker: Anand Iyer, University of California, Berkeley Talk Title: Scalable Systems for Large-Scale Dynamic Connected Data Processing Series: CS Colloquium Abstract: As the proliferation of sensors rapidly make the Internet-of-Things (IoT) a reality, the devices and sensors in this ecosystem-”such as smartphones, video cameras, home automation systems and autonomous vehicles-”constantly map out the real-world producing unprecedented amounts of connected data that captures complex and diverse relations. Unfortunately, existing big data processing and machine learning frameworks are ill-suited for analyzing such dynamic connected data, and face several challenges when employed for this purpose. In this talk, I will present my research that focuses on building scalable systems for dynamic connected data processing. I will discuss simple abstractions that make it easy to operate on such data, efficient data structures for state management, and computation models that reduce redundant work. I will also describe how bridging theory and practice with algorithms and techniques that leverage approximation and streaming theory can significantly speed up computations. The systems I have built achieve more than an order of magnitude improvement over the state-of-the-art and are currently under evaluation in the industry for real-world deployments. This lecture satisfies requirements for CSCI 591: Research Colloquium. Biography: Anand Iyer is a PhD candidate at the University of California, Berkeley advised by Prof. Ion Stoica. His research interest is in systems with a current focus on enabling efficient analysis and machine learning on large-scale dynamic, connected data. He is a recipient of the Best Paper Award at SIGMOD GRADES-NDA 2018 for his work on approximate graph analytics. Before coming to Berkeley, he was a member of the Mobility, Networking and Systems group at Microsoft Research India. He completed his M.S at the University of Texas at Austin. Host: Barath
          Senior Java Developer - £60k - £80k - Burgess Hill, West Sussex      Cache   Translate Page      
Senior Java Developer - 60k - 80k - Burgess Hill, West Sussex A Senior Java Developer is required by a leading company based in Burgess Hill, West Sussex. They are currently going through a large expansion phase and looking for a Senior Java Developer to add to their current software development team. The position will involve working in small sub-teams within a larger development team working in an agile environment. Essential experience: Java 7 or 8 Spring Restful APIs Agile Any experience in one or more of the following would be advantageous: Angular, React Git, Maven, Jenkins Junit TDD Machine learning Splunk Selenium This is an opportunity to work in an R&D environment using the latest technology stack on varied Greenfield projects. The company are advocates of a good work/life balance and offer a strong benefits package. If you are looking for an opportunity of this nature please contact or call .
          Master Data Management Market Covering Trends, Market Share and Forecast to 2026      Cache   Translate Page      

Report on the global master data management market provides analysis for the period between 2016 and 2026, wherein 2017 is the base year, and 2018 to 2026

Albany, NY -- (SBWIRE) -- 03/11/2019 -- Market Research Reports Search Engine (MRRSE) has been serving as an active source to cater intelligent research report to enlighten both readers and investors. This research study titled "Master Data Management Market "

Get Report Sample Copy https://www.mrrse.com/sample/1829

Backed by pressing needs for consolidated data management and advanced analytics to ensure seamless business workflow and data consistency have augmented adoption of advanced master data management. Large scale adoption by IT industries in the face of rampant technological advances such as IoT and corresponding rise in data accumulation further create ample growth opportunities in master data management market. Such market highlights are in sync with Market Research Reports Search Engine's (MRRSE) recent report addition titled, 'Master Data Management Market – Global Industry Analysis, Size, Share, Growth, Trends, and Forecast 2018 – 2026' included in its fast expanding online data archive.

To remain viable choices amongst amidst staggering competition, leading players are keen on solution improvisation to remain industry specific. Inclusion of advances features such as AI and machine learning are likely to further enhance capabilities of master data management services, thereby augmenting ample growth upsurge.

This elaborate research report on master data management market commences with a detailed executive summary encompassing market definition and dynamics followed by an overview section and market dynamics such as drivers, restraints, threats, and challenges as well as other macro and micro economic factors that amplify growth in master data management market.

To View Complete Research Report with TOC @ https://www.mrrse.com/master-data-management-market

The report also includes a detailed section on market segmentation on the basis of which master data management market is classified by solution, deployment type, enterprise size, and industry. By solution master data management market is split into product, customer, supplier, and multi-domain MDM solutions. In terms of deployment type master data management market is bifurcated into on-premise and cloud based. By enterprise size, the market is further classified into large and small and medium enterprises. In terms of industry master data management market is also bifurcated into BFSI, IT and telecom, healthcare, retail, and manufacturing amongst others.

Further in the course of the report readers are also presented with substantial insights on regional scope and diversification on the basis of which readers can effectively identify potential lucrative regions in master data management market. By region the market is demarcated into Europe, North and South America, Asia-Pacific, and MEA.

Amidst staggering competition, leading players in master data management market are fast employing profit oriented winning strategies to secure their leading stance. The report in its trailing sections offers tangible information on the efficacy of marketing strategies in steering favorable progress in master data management market. The report includes detailed analysis as well as SWOT analysis on potential leading market players and their winning strategies, evaluating their efficacy in ensuring sustainable revenue pools.

Send an Enquiry @ https://www.mrrse.com/enquiry/1829

About Market Research Reports Search Engine (MRRSE)
Market Research Reports Search Engine (MRRSE) is an industry-leading database of Market Research Reports. MRRSE is driven by a stellar team of research experts and advisors trained to offer objective advice. Our sophisticated search algorithm returns results based on the report title, geographical region, publisher, or other keywords.

MRRSE partners exclusively with leading global publishers to provide clients single-point access to top-of-the-line market research. MRRSE's repository is updated every day to keep its clients ahead of the next new trend in market research, be it competitive intelligence, product or service trends or strategic consulting.

Contact Us
State Tower
90, State Street
Suite 700
Albany, NY - 12207
United States Telephone: +1-518-730-0559
Email: sales@mrrse.com

For more information on this press release visit: http://www.sbwire.com/press-releases/master-data-management-market-covering-trends-market-share-and-forecast-to-2026-1167313.htm

Media Relations Contact

alok
Manager
MRRSE
Telephone: 1-518-621-2074
Email: Click to Email alok
Web: https://www.mrrse.com/

#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000


          A Personalized Account-Based Recommender within the University Library’s Mobile App Interface      Cache   Translate Page      
A Personalized Account-Based Recommender within the University Library’s Mobile App Interface Hahn, James F. With funding from the University of Illinois Campus Research Board, researchers developed a personalized account-based recommender within the university library’s mobile app interface. The recommender system (RS) is derived from two key data sources 1) data mining of item topic clusters checked out together in the university library and 2) consequent topics from data mining paired with filtered searches of the library catalog. Machine Learning; IMLS; personalization; recommender system; Weka
          Senior AI/Deep Learning Software Engineer - St Josephs Hospital and Medical Center - Phoenix, AZ      Cache   Translate Page      
Ability to align business needs to development and machine learning or artificial intelligence solutions. Experience in natural language understanding, computer...
From Dignity Health - Tue, 27 Nov 2018 03:06:49 GMT - View all Phoenix, AZ jobs
          Offer - PRACTICAL SSRS 2017 Online Training @ SQL School - AUSTRIA      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. SQL Server Reporting Services : Features of our Training: • Completely Practical • Completely Real time • Highly Interactive • Real time Case Studies • Interview Guidance • Certification Guidance • Mock Interviews • Job Support All Sessions are Completely Practical and Realtime. For free SSRS Online Demo, please visit : http://sqlschool.com/SSRS-Online-Training.html Schedules for PRACTICAL SQL 2016 & 2017 SSRS Online TRAINING : http://sqlschool.com/Register.html Contact us today (24 x 7) for SSRS Practical Online Training SQL School Training Institute ISO 9001:2008 Certified Organization for Training Authorized Microsoft Partner (ID# 5108842) India: Mobile: +91 (0) 9666 44 0801 Mobile: +91 (0) 9666 64 0801 USA: Office: +1 (510) 400-4845 Office 1: #101, UMA Residency, Opp: Sindhu Travels, Beside Metro Station Gate #D, SR Nagar, Hyderabad - 38, India. Office 2: #202, Sai Anu Avenue, Street #3, Patrika Nagar, HITECH City, Hyderabad -81, India. Website: http://sqlschool.com/ Follow us: https://www.facebook.com/sequelschool https://www.linkedin.com/company/sql-school https://twitter.com/sequelschool
          Nubank contrata ex-vice da IBM para serviço de dados      Cache   Translate Page      

Venkatraman tem 20 anos de experiência em algoritmos e machine learning [...]

O post Nubank contrata ex-vice da IBM para serviço de dados apareceu primeiro em Forbes Brasil.


          Quand les robots sucent le cerveau des auteurs      Cache   Translate Page      

Pendant longtemps, « Don’t be evil » (qu’on pourrait traduire par « Ne faites pas le mal ») a été le mantra de Google – sa devise informelle. Remplacée en 2018 par « Do the right Thing » (« Faites ce qui est bien »), elle synthétise pourtant l’époque d’une manière aussi clairvoyante que concise : chaque innovation pourra être utilisée à bon ou à mauvais escient.


Par Julien Simon
de EPUBNerd


Love Robot
Alex Guibord, CC BY ND 2.0

 

On passera sur les implications philosophiques desdites devises – qu’est-ce que le Mal dans le domaine de l’ingénierie, d’ailleurs qu’est-ce que le Mal tout court ? – pour se concentrer sur l’ambivalence permanente dans laquelle nous flottons : la plupart des technologies que nous utilisons de façon quotidienne peuvent aussi bien être utilisée à notre avantage qu’en notre défaveur.
 

Prenons un exemple : une enceinte connectée type Alexa (Amazon) est capable, d’un simple ordre vocal, de lancer la lecture d’une playlist, d’allumer la lumière ou de lire une histoire à votre enfant. Mais cette même enceinte peut aussi poser de graves problèmes de sécurité (en étant hackée) ou de confidentialité (vous avez laissé Amazon poser un micro dans votre salon, tout de même).
 

Le cas du livre numérique


De la même manière, l’émergence du livre numérique et des plateformes d’autopublication a permis à des millions d’auteurs de diffuser leurs créations à grande échelle, de façon dématérialisée mais aussi imprimée (grâce à l’impression à la demande). Certains d’entre eux ont bâti de véritables notoriétés, et sans aller jusqu’à parler de fortune, à gagner plus d’argent que ne leur aurait permis un contrat d’édition « classique ».
 

Mais cette disponibilité a un coût, autant financier que moral : celui des infrastructures qui mettent ces livres à disposition. Aujourd’hui, Amazon et Google (Apple et Kobo également, dans une moindre mesure) assurent l’essentiel de ce trafic. Ces entreprises s’assurent que les livres dont vous leur confiez les fichiers soient disponibles à tout moment : vous bénéficiez de la puissance de leur infrastructure. En échange, ces distributeurs ponctionnent leur part sur chaque transaction effectuée – les petits ruisseaux faisant les grandes rivières.
 

Mais est-ce bien tout ce qu’ils ponctionnent ?


Car tout le monde pourrait être content… si Amazon, Google et consorts ne mettaient pas autant d’énergie à développer des programmes d’intelligence artificielle basés sur le machine learning. Et ils ne sont pas les seuls : il y a quelques semaines, la plateforme d'autopublication Wattpad présentait son algorithme destiné à aider sa future maison d’édition à dénicher les bestsellers.

Comme Amazon et Google (pourquoi croyez-vous d’ailleurs que Google se soit lancé dans un programme de numérisation massive des fonds de bibliothèques américaines), Wattpad est dans une position idéale pour cela, puisqu’elle dispose d’un grand nombre de manuscrits publiés sur sa plateforme gratuite et que le machine learning nécessite justement l’analyse d’un très grand nombre de documents afin d’y repérer des schémas.
 

En bref : les plateformes auxquelles les auteurs confient leurs précieux manuscrits sont aussi celles qui sont les plus actives dans le domaine de l’intelligence artificielle et du machine learning. Vous voyez où je veux en venir ?

 

Les algorithmes au clavier

 

Aujourd’hui, en salle de presse, il n’est plus tabou de parler d’algorithme rédacteur de contenu. Ainsi, le Los Angeles Times a conçu un quake-bot qui rédige et publie automatiquement un article à chaque tremblement de terre. D’autres quotidiens, papier et web, utilisent l’intelligence artificielle pour mettre en avant les résultats sportifs ou les bulletins météo. La France n’est pas en reste : en 2015, Le Monde utilisait déjà un programme chargé de publier une page web pour chacune des 36.000 municipalités afin d’afficher le résultat des élections (source : Le Devoir).

Et il n’y a pas que la presse à être touchée : en 2016, le programme Benjamin écrivait le scénario d’un court-métrage de science-fiction. Et même si le résultat était un peu… étonnant, il n’en demeurait pas moins impressionnant.



 

Le remplacement des auteurs n’est plus du domaine de la science-fiction

 

L’intelligence artificielle brûle de raconter des histoires. Et de là à toucher, d’ici à quelques années, aux domaines de la littérature, il n’y a qu’un pas. Mais pour cela, il faudra bien entendu que les machines s’entraînent. Et il leur faudra lire beaucoup, beaucoup, d’histoires, beaucoup, beaucoup, de manuscrits, analyser leurs ressorts narratifs, mais aussi étudier ce qui fonctionne ou pas, quel texte remporte les faveurs des utilisateurs, à quel moment la lectrice A cesse sa lecture et à quel moment le lecteur B l’interrompt, afin d’en dégager des « patterns » .

Et pour cela, le meilleur moyen est encore de posséder toutes les infrastructures d’une librairie en ligne et d’un service de lecture numérique.
 

On aboutit alors à une situation paradoxale qui, avouons-le, ne manque pas de piquant : ce sont les écrivains et les éditeurs d’aujourd’hui qui, en publiant leurs textes sur ces plateformes, nourrissent les machines qui, à terme, apprendront à les remplacer. Car plus on publiera sur Amazon, Google, Wattpad, etc, meilleurs deviendront leurs algorithmes, et plus vite on accélèrera l’avènement d’un tel futur.
 

Et pourtant, si l’on veut s’autopublier – et si l’on veut même simplement vendre des livres numériques, en tant qu’indépendant ou industriel –, ces plateformes demeurent les plus performantes. Elles sont aussi et surtout les plus riches du point de vue de la masse d’utilisateurs déjà captifs de leur écosystème. D’où le dilemme. Car dans les conditions d’utilisation de Kindle Direct Publishing, rien n’explicite clairement qu’un tel usage pourra être fait de nos histoires… mais rien ne l’interdit non plus (voir paragraphe 5.5 : Concession de droits).
 

Don’t be evil, do the right thing… Finalement tout cela dépend de nos propres notions de bien et de mal, et de notre propre tolérance à la servitude volontaire. Cela dépend aussi des calculs stratégiques que nous effectuons à l’échelle de l’industrie.
 

En attendant, les machines apprennent.


Dossier : L'intelligence artificielle au service du livre et de la lecture


          Datanauts 121: A Professor Takes Us To Machine Learning School      Cache   Translate Page      

The term 'Machine Learning' is being sprinkled over IT marketing materials like magic dust. The Datanauts get to what’s real in ML with guest Vivian Zhang.

The post Datanauts 121: A Professor Takes Us To Machine Learning School appeared first on Packet Pushers.


          Neuroscientists can read brain activity to predict decisions 11 seconds before people act      Cache   Translate Page      
Neuroscientists can read brain activity to predict decisions 11 seconds before people act:

Free will, from a neuroscience perspective, can look like quite quaint. In a study published this week in the journal Scientific Reports, researchers in Australia were able to predict basic choices participants made 11 seconds before they consciously declared their decisions.

In the study, 14 participants—each placed in an fMRI machine—were shown two patterns, one of red horizontal stripes and one of green vertical stripes. They were given a maximum of 20 seconds to choose between them. Once they’d made a decision, they pressed a button and had 10 seconds to visualize the pattern as hard as they could. Finally, they were asked “what did you imagine?” and “how vivid was it?” They answered these questions by pressing buttons.

Using the fMRI to monitor brain activity and machine learning to analyze the neuroimages, the researchers were able to predict which pattern participants would choose up to 11 seconds before they consciously made the decision. And they were able to predict how vividly the participants would be able to envisage it.

Lead author Joel Pearson, cognitive neuroscience professor at the University of South Wales in Australia, said that the study suggests traces of thoughts exist unconsciously before they become conscious. “We believe that when we are faced with the choice between two or more options of what to think about, non-conscious traces of the thoughts are there already, a bit like unconscious hallucinations,” he said in a statement. “As the decision of what to think about is made, executive areas of the brain choose the thought-trace which is stronger. In, other words, if any pre-existing brain activity matches one of your choices, then your brain will be more likely to pick that option as it gets boosted by the pre-existing brain activity.”

The work has implications for how we understand uncomfortable thoughts: Pearson believes the findings explain why thinking about something only leads to more thoughts on the subject, as it creates “a positive feedback loop.” The study also suggests that unwelcome visualizations, such as those experienced with post-traumatic stress disorder, begin as unconscious thoughts.



          Principal Security Data Scientist - Sierra Nevada Corporation - Sparks, NV      Cache   Translate Page      
DATA SCIENCE / MACHINE LEARNING SKILLS:. Sierra Nevada Corporation is an Equal Opportunity Employer. Required to act as a trusted adviser for business leaders...
From Sierra Nevada Corporation - Fri, 15 Feb 2019 23:07:32 GMT - View all Sparks, NV jobs
          Machine Learning : les différentes manières dont le « as a Service » démocratise l'IA      Cache   Translate Page      
none
          Data Engineer (Data Warehouse) - Bandwidth - Raleigh, NC      Cache   Translate Page      
Machine Learning & statistics experience. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Mon, 25 Feb 2019 20:43:17 GMT - View all Raleigh, NC jobs
          Senior Software Engineer, Network Engineering - Bandwidth - Raleigh, NC      Cache   Translate Page      
You've got experience with machine learning and anomaly detection. 90-minute fitness lunch with a paid gym membership with shuttle service available for...
From Bandwidth - Tue, 19 Feb 2019 16:35:34 GMT - View all Raleigh, NC jobs
          Autonomous Systems Manager - DOT Technology Corp. - Emerald Park, SK      Cache   Translate Page      
Interest and/or experience with AI, machine learning, robotics, and autonomous vehicles. DOT Technology Crop....
From Indeed - Fri, 15 Feb 2019 04:13:34 GMT - View all Emerald Park, SK jobs
          Full Stack Software Developer - DOT Technology Corp - Emerald Park, SK      Cache   Translate Page      
Interest and/or experience with AI, machine learning, and robotics. Welcome to DOT Technology Corp!...
From Indeed - Thu, 31 Jan 2019 21:41:48 GMT - View all Emerald Park, SK jobs
          [Upwork] LinkedIn Web Crawler Expert for Lead Acquisition      Cache   Translate Page      
From Upwork // CoachList (www.coachlist.com) is a new online marketplace platform for sports & fitness athletes and enthusiasts. CoachList is a transactional & communications platform where athletes can search for a sports service provider, filter through qualifications & reviews, instantly book training sessions, and communicate using our messaging app. Sports providers can use our CoachList Pro SaaS products and business management tools to optimize training session revenues, transact business and take advantage of new eCommerce & media channels to grow & market their services.

CoachList's marketplace enables communication and transactions between consumers (athletes, fitness enthusiasts, weekend warriors, teams, parents, etc.) and sports & fitness industry service providers. CoachList's products help sports consumers to identify trainers, instructors, fitness facilities, camps, events & non-training service providers and to manage their training. For the sports & fitness industry service providers our products and distribution channels create an affordable and effective way to acquire new local in-person & global online customers and to optimize revenues.

CoachList has a proven executive team a team that has the deep expertise in aggregating fragmented online marketplaces globally across all languages and markets. It understands how to aggregate supply and demand ecosystems from fragmented user bases. They also have deep expertise in natural search, paid search, lead nurturing, marketing automation, plus customer acquisition and engagement, with the ability to engage our user base emotionally, so our tools delight users, and feel indispensable. CoachList uses industry leading engagement and conversion strategies, including gamification, behavioral modeling, multivariate testing, and world-class user experiences. CoachList leverages a network of best-in-class partners to decrease cost and increase expertise in non-core areas, and leverages its own intellectual property to differentiate from competitors in core parts of its business. In addition, its advisor network of professional athlete luminaries serve to expand visibility and to serve as advisors, it’s low-cost of operations, superior technology, and lean development model poise CoachList to become highly profitable in this growing yet untapped market.

The duties and responsibilities of a CoachList Web Crawler, Bot, and Process Automation Engineer are:
• Develop web crawlers to crawl search engines and other databases for qualified leads
• Apply machine learning to classify, categorize, rank, and qualify leads
• Use Machine learning to find new lead sources
• Develop custom Selenium applications to automate manual lead acquisition processes
• Keep abreast with the latest ML, Process Automation, and Web Crawling technologies.
• Research and Identify new data sources and keywords to expand the lead acquisition efforts of CoachList.
• Assist other employees in automating their manual processes


Posted On: March 12, 2019 22:02 UTC
Category: Sales & Marketing > Lead Generation
Country: United States
click to apply
          Computer Scientist, GS-1550-09/11 (DEU-DP) - US Department of the Interior - Middleton, WI      Cache   Translate Page      
Researches existing machine learning techniques, and applies appropriate methods towards prediction. For additional information on our internal telework policy,... $50,598 - $79,586 a year
From usajobs.gov - Mon, 04 Mar 2019 10:08:38 GMT - View all Middleton, WI jobs
          Dynamics CRM Developer - Concurrency - Brookfield, WI      Cache   Translate Page      
Combining two decades of CRM and ERP developments with innovative technology such as machine learning and automation, Dynamics 365 is helping businesses around...
From Concurrency - Wed, 30 Jan 2019 20:34:33 GMT - View all Brookfield, WI jobs
          Microsoft Dynamics 365 Engineer - Concurrency - Brookfield, WI      Cache   Translate Page      
Combining two decades of CRM and ERP developments with innovative technology such as machine learning and automation, Dynamics 365 is helping businesses around...
From Concurrency - Wed, 26 Dec 2018 20:30:31 GMT - View all Brookfield, WI jobs
          machineByte: Reporter - Machine Learning, machineByte - NY / IL      Cache   Translate Page      
Competitive: machineByte: machineByte, an online news site dedicated to machine learning and data science in finance, is looking for a reporter to cover the Americas region. New York City or Chicago
           It takes more brain power to forget something than to remember it       Cache   Translate Page      

The team used fMRI and machine learning to track brain function#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

An important part of the human brain has to work harder to actively forget a memory than it does to remember it, according to the results of a newly-published study. The research is a step towards understanding how and why the brain is able to discard an experience, and could one day lead to a treatment designed to remove painful memories.

.. Continue Reading It takes more brain power to forget something than to remember it

Category: Science

Tags:
          Roasted corn — and machine learning in a food truck — at 23rd and Union’s El Costeño      Cache   Translate Page      
Moises Santos is a 24-year-old programmer, food truck designer, and immigrant from Oaxaca, Mexico. His food truck holds down what seems like prime territory — the pot purchasing and stoner friendly parking lot at the Central District’s Uncle Ike’s. The … Continue reading
          Assessing Causality from Observational Data using Pearl’s Structural Causal Models      Cache   Translate Page      
Causality In 20th century statistics classes, it was common to hear the statement: “You can never prove causality.” As a result, researchers published results saying “x is associated with y” as a way of circumventing the issue of causality yet implicitly suggesting that the association is causal. As an example from my former discipline, political science, there was an interest in determining how representative democracy works. Do politicians respond to voters, or do voters just update their policy beliefs to line up with the party they’ve always preferred? It turns out that this is a very difficult question to answer, so political scientists interested in publishing choose their language carefully and pronounce that policy “congruence” exists between voters and politicians. The upshot is that there now exists a scholarly literature on “voter-party congruence,” which tells you exactly nothing about how democracy works but allows democracy researchers to get their papers past peer review. 21st century understandings of causality, however, have evolved away from 20th century fatalism to reframe the question as: What assumptions need to be met in order to state that an association is causal? Under what conditions are those assumptions met? Can these assumptions be met even when we can’t perform randomization? There are two conceptually different approaches to the problem: Donald Rubin’s (elaboration on Jerzy Neyman’s) potential outcomes framework. Judea Pearl’s (elaboration on Sewall Wright’s) structural causal models (SCMs). The former is the dominant approach in applied statistics, but the latter approach can sometimes highlight unexpected results that inform the proper analysis of observational data. Before describing the SCM framework, the next section reviews the potential outcomes framework. Potential Outcomes Take a binary treatment \(D_i \in \{0, 1\}\). Represent the outcome received by subject i as \(Y_{iD}\). Then \(Y_{i0}\) and \(Y_{i1}\) are the potential outcomes. A subject is either \(Y_{i0}\) or \(Y_{i1}\), we don’t observe both. Yet we want to determine: \(Y_{i1} - Y_{i0}\) which is the causal effect of the intervention. Although subjects receive either 0 or 1, but not the other, we may be able to identify the Average Treatment Effect (ATE). \(\text{ATE} = \mathbb{E}\left[Y_{i1} - Y_{i0}\right]\) To derive appropriate estimators for the ATE we need to make a few assumptions. Particularly important is that the treatment is independent of potential outcomes, written as: \(Y_{i0}, Y_{i1} \perp\!\!\!\perp D_i\) Finding ways to make \(D_i\) independent is at the heart of the potential outcomes framework. This leads to a few methodologies now commonplace in applied statistics: Randomized experiments by definition make \(D_i\) independent. Propensity score matching or weighting make the treated and controls look the same on possible confounders so that the only differences must be random error. Regression discontinuity designs where a cut-off on a continuous variable separates treated and control units. Instrumental variables, where compliance is non-random but treatment assignment is random. Longitudinal designs that use fixed effects or first differences to remove unit-level confounders affecting the treatment. The key assumption is \(Y_{i0}, Y_{i1} \perp\!\!\!\perp D_i\), termed ignorability. Judea Pearl has criticized how unintuitive the potential outcomes framework makes this assumption. He writes in The Book of Why (2018, pg. 279-280): “Unfortunately, I have yet to find a single person who can explain what ignorability means in a language spoken by those who need to make this assumption or assess its plausibility in a given problem…If you think this sounds circular, I agree with you!” Instead, Pearl has spent the last twenty years developing a different orientation that builds off his work in the 1990s on Bayesian networks. (It’s not necessary to know how Bayesian networks work to understand this post, but it does help clarify how his thinking on the problem of causality evolved over several decades.) This orientation is known as structural causal models (SCMs). Structural Causal Models SCMs are graphs with nodes, directed edges, and functions mapping exogenous variables to endogenous ones. Denote \(U\) as the set of exogenous variables, \(V\) as the set of endogenous variables, and \(F\) as the set of functions mapping \(U\) to \(V\). A concrete example is: \(U = \{X, Y\}\) \(V = \{Z\}\) \(F = \{f_z\}\) where \(f_z\) is the function mapping \(X\) and \(Y\) onto \(Z\). This definition implies the following graph: The arrows represent a generic causal relationship only, the actual function mapping \(X\) and \(Y\) onto \(Z\) can be anything we like. These types of figures should be familiar to anybody who has previously encountered structural equation models (SEMs) in applied statistics. The primary difference is that SEMs are parametric, typically assuming a linear relationship: \(Z = b_0 + b_1X + b_2Y\) but SCMs are defined without committing to a particular functional form. We get around functional forms by talking about the variables in terms of joint probability functions and taking advantage of well-known rules for converting between joint, conditional, and marginal probabilities. Take the following graph: Any (acyclic) graph has a joint distribution that is defined by multiplying all conditional probabilities, where conditioning is performed on the direct parent. For example, the joint distribution for the variables in the model is \(P(X, Y, Z) = P(X)P(Y \vert X)P(Z \vert Y)\) Understanding the conditional probabilities implied by a model will enable us to generate some rules for determining how causal effects can be identified from observational data. These rules provide surprising and important perspectives on how statistical modeling should be approached. Backdoor Paths and Colliders “You should control for everything you can. That is, after all, why we do regression.” - One of my methodology professors in the early 2000s. No, you should not control for everything. In fact, depending on the causal model, some variables should explicitly not be controlled for. We’ll start out with when you should control for a non-treatment variable. Take the following graph: We wish to know the effect of \(X\) on \(Z\), but \(Y\) is a common cause. Let’s say we could intervene in the world to set \(X\) at a given value. By doing so, we’d be removing the effect of \(Y\) on \(X\) and would be left with: We can identify the causal effect by comparing the world in which we have control with the world in which we do not. In both scenarios, the probability that \(Z\) takes on a value is conditioned only on \(Y\) and \(X\), \(P(Z = z \mid Y, X)\), and the probability that \(Y\) takes on a given value is not conditional on anything. We want to know the effect of \(X\) on \(Z\) if we could intervene on \(X\) and set its value. Pearl introduces the \(do(\cdot)\) operator to signify setting a variable \(X\) to a specific value \(x\). \(P(Z = z \mid do(X = x))\) Based on the intervention SCM, \(P(Z = z \mid do(X = x)) = \sum_z P(Z = z \mid Y = y, X = x)P(Y = y)\) This is true because \(P(Z = z \mid do(X = x))\) is what we get after integrating out \(Y\). But we know from comparing the graphs that \(P(Z = z \mid Y = y, X = x)\) and \(P(Y = y)\) are the same in both worlds. Thus, we have all the information we need to calculate a causal effect such as \(P(Z = z \mid do(X = 1)) - P(Z = z \mid do(X = 0))\) Take a slightly more complicated model: There are now two paths from \(X\) to \(Z\): \(X \rightarrow Z\) \(X \leftarrow W \rightarrow Y \rightarrow Z\) These are read from left to right regardless of the direction of the arrows. However, the arrows identify the second path as a backdoor path because there is an arrow leading into \(X\). Backdoor paths are essential for identifying causal effects because they represent spurious associations. Pearl shows that causal effects can be identified if we can block the backdoor path. We do this by conditioning on any of the variables that lay on the backdoor path, meaning the conditioning set can be any of the following: \(\{W\}\) \(\{Y\}\) \(\{W, Y\}\) We don’t necessarily have to control for both, though we can. The key is that, by blocking a backdoor path, we remove the spurious association between the outcome and \(X\). After blocking, we do not necessarily need to control for subsequent variables on the backdoor path. Now let’s flip the top arrows. This fundamentally changes the conditioning set, which now only contains \(Y\). This occurs because \(W\) is a collider variable, which is defined as a variable that lies along a backdoor path with arrows pointing into it from multiple directions. We would write this backdoor path as \(X \rightarrow W \leftarrow Y \rightarrow Z\). When we write out the path in this manner, we can immediately identify collider variables as those with arrows pointing to the node from both directions. A collider variable blocks a backdoor path. The counter-intuitive result is that conditioning on a collider opens the backdoor path. To identify the causal effect we need to block all backdoor paths from \(X\) to \(Z\). The backdoor criterion can be defined as (Pearl, Glymour, & Powell, 2016, p. 61): Given an ordered pair of variables \((X,Z)\) in a directed acyclic graph \(G\), a set of variables \(V\) satisfies the backdoor criterion relative to \((X,Z)\) if no node in \(V\) is a descendant of \(X\), and \(V\) blocks every path between \(X\) and \(Z\) that contains an arrow into \(X\). That is, we identify a set of nodes in \(\{V\}\) to condition on such that: We block all spurious paths from \(X\) to \(Z\). We leave all directed paths from \(X\) to \(Z\) unperturbed. We do not inadvertantly create new spurious paths via conditioning on colliders or their descendants. Mediation Another example is mediation, as in the following figure: We can get the direct effect of \(X\) on \(Z\) if we average over levels of \(M\), which is the standard approach to mediation. But what if we add a variable as follows?: Now \(M\) is a collider, and we know that conditioning on a collider causes problems. Conditioning on \(M\) opens the path \(X \rightarrow M \leftarrow W \rightarrow Z\), allowing an indirect effect to interfere with the direct effect. But not conditioning on \(M\) leaves the indirect path \(X \rightarrow M \rightarrow Z\) open. How do we deal with this in a manner that allows us to recover the direct effect of \(X\) on \(Z\)? The answer is that we now intervene on both \(X\) and \(M\). \(P(Z=z \mid do(X = x), do(M = m))\). Intervening and conditioning are not the same thing. Conditioning averages over values of \(M\), intervening sets its value such that there are no longer the arrows \(X \rightarrow M\) and \(W \rightarrow M\). The conditional direct effect is \(CDE = P(Z=z \mid do(X = x), do(M = m)) - P(Z=z \mid do(X = x^{\prime}), do(M = m))\) The conditional refers to the fact that the direct effect \(X \rightarrow Z\) may differ depending on the value to which the mediator is set. The \(do(\cdot)\) operator is equivalent to removing an arrow from a graph. Reiterating the model: There is no path to \(X\), so \(do(X) = x\), and the CDE is \(CDE = P(Z=z \mid X = x, do(M = m)) - P(Z=z \mid X = x^{\prime}, do(M = m))\). The last step is to rewrite the \(do(M = m)\) in terms of the observed world. To block the backdoor path \(M \leftarrow W \rightarrow Z\) we need to condition on \(W\). We are left with: \[\begin{eqnarray} CDE = \sum_i \left[P(Z=z \mid X = x, M = m, W = w) - \\ P(Z=z \mid X = x^{\prime}, M = m, W = w)\right]P(W = w) \end{eqnarray}\] There is a general result behind this (Pearl, Glymour, & Jewell, 2016, pg. 77): The CDE of \(X\) on \(Z\) can be identified when a mediation variable \(M\) is present given: There exists a set \(V_1\) of variables that blocks all backdoor paths from \(M\) to \(Z\). There exists a set \(V_2\) of variables that blocks all backdoor paths from \(X\) to \(Z\) after deleting all arrows entering \(M\). The second of these was met automatically given the lack of parents for \(X\). These general rules make it possible to identify direct causal effects in contexts that were previously intractable, even if the researchers did not realize they were dealing with an intractable problem. The daggity R Package These models are all very simple, but graphs can be far more complex. Consider the following (adapted from Morgan & Winship, 2015, pg. 135): A general approach to modeling these diagrams is to employ a tool called d-separation, defined as follows (Pearl, Glymour, & Powell, 2016, p. 47): A path \(p\) is blocked by a set of nodes \(N\) iif: \(p\) contains a chain of nodes \(A \rightarrow B \rightarrow C\) or fork \(A \leftarrow B \rightarrow C\) such that the middle node \(B\) is conditioned on, or \(p\) contains a collider \(A \rightarrow B \leftarrow C\) such that the collision node \(B\) is not conditioned on, nor are any descendents of \(B\) conditioned on. Fortunately, there is software that can help us algorithmically determine which variables are d-separated. The software (and R package) is called dagitty. To use the package, we start by declaring the SCM: g % head(15) ## { S, V } ## { S, T, V } ## { S, U, V } ## { T, U, V } ## { S, T, U, V } ## { S, V, W } ## { S, T, V, W } ## { U, V, W } ## { S, U, V, W } ## { T, U, V, W } ## { S, T, U, V, W } ## { S, V, Y } ## { T, V, Y } ## { S, T, V, Y } ## { S, U, V, Y } Notice that \(T\) is in some of these sets. If we unblock the path \(X \leftarrow S \rightarrow T \leftarrow U \rightarrow Y \rightarrow Z\), we need to reblock it by conditioning on another variable such as \(U\) or \(Y\). This is a lot of options. Can we get something simpler? adjustmentSets(g, "X", "Z", type = "minimal") ## { V, W, Y } ## { T, V, Y } ## { U, V, W } ## { T, U, V } ## { S, V } Note two important points. We don’t have to condition on all possible causes of \(Y\). There are some combinations of variables we should not use as adjustors. We’ll illustrate by generating some data consistent with the model. The SEM package lavaan makes generating data for simultaneous equations relatively easy. lavaan_model % tidy() ## # A tibble: 5 x 5 ## term estimate std.error statistic p.value ## ## 1 (Intercept) 0.0227 0.0322 0.704 4.81e- 1 ## 2 X 0.801 0.0278 28.8 3.81e-133 ## 3 V 0.602 0.0322 18.7 4.64e- 67 ## 4 W 0.574 0.0298 19.3 1.25e- 70 ## 5 Y 0.582 0.0298 19.5 5.87e- 72 Model 3: lm(Z ~ X + U + V + W, data = g_tbl) %__% tidy() ## # A tibble: 5 x 5 ## term estimate std.error statistic p.value ## ## 1 (Intercept) 0.0568 0.0364 1.56 1.19e- 1 ## 2 X 0.800 0.0316 25.3 1.49e-109 ## 3 U 0.337 0.0384 8.77 7.37e- 18 ## 4 V 0.585 0.0371 15.8 3.62e- 50 ## 5 W 0.557 0.0344 16.2 1.67e- 52 Model 4: lm(Z ~ X + T + U + V, data = g_tbl) %__% tidy() ## # A tibble: 5 x 5 ## term estimate std.error statistic p.value ## ## 1 (Intercept) 0.0657 0.0403 1.63 1.03e- 1 ## 2 X 0.820 0.0356 23.0 1.63e-94 ## 3 T 0.248 0.0427 5.80 9.10e- 9 ## 4 U 0.384 0.0442 8.68 1.60e-17 ## 5 V 0.580 0.0430 13.5 3.48e-38 We get much closer to the true causal effect estimate whenever we use the conditioning sets suggested by daggity. Unobservable or Unmeasurable Variables Once again, take our model: Let’s say that we can’t actually observe \(W\) or \(Y\). An old-school regressionista would say we are SOL. A modern causal-aware practitioner would not. We can tell dagitty that these variables are unobserved, or latent. g_unobs % mutate_if(is.numeric, funs(round(., 3))) ## Warning: funs() is soft deprecated as of dplyr 0.8.0 ## please use list() instead ## ## # Before: ## funs(name = f(.) ## ## # After: ## list(name = ~f(.)) ## This warning is displayed once per session. ## # A tibble: 2 x 5 ## term estimate std.error statistic p.value ## ## 1 (Intercept) 0.017 0.035 0.485 0.628 ## 2 V 0.258 0.029 8.81 0 If our model is correct, controlling for \(T\) should render this association statistically indistinguishable from zero. Does it? lm(W ~ V + T, data = g_tbl) %__% tidy() %__% mutate_if(is.numeric, funs(round(., 3))) ## # A tibble: 3 x 5 ## term estimate std.error statistic p.value ## ## 1 (Intercept) 0.016 0.031 0.522 0.602 ## 2 V 0.009 0.03 0.288 0.774 ## 3 T 0.485 0.03 16.2 0 In fact, we can get all conditional independencies implied by the model. impliedConditionalIndependencies(g) %__% head(20) ## S _||_ U ## S _||_ V | T ## S _||_ W | T ## S _||_ Y ## S _||_ Z | V, W, X, Y ## S _||_ Z | T, V, X, Y ## S _||_ Z | U, V, W, X ## S _||_ Z | T, U, V, X ## T _||_ X | S, V ## T _||_ Y | U ## T _||_ Z | V, W, X, Y ## T _||_ Z | U, V, W, X ## T _||_ Z | S, V, W, Y ## T _||_ Z | S, U, V, W ## U _||_ V | T ## U _||_ W | T ## U _||_ X | S, V ## U _||_ X | S, T ## U _||_ Z | V, W, X, Y ## U _||_ Z | S, V, W, Y We generated our data to intentionally be consistent with the model, so testing these conditional independencies will confirm them. When we don’t know if the model is correct, however, we can generate the conditional independencies and check each of them. If they are not correct, our model is wrong. When \(\{V\}\) is large, the possible set of connections may not all be clearly dictated by theory, and the number of possible combinations of arrows is too large to test via a grid-search. Familiarity with Pearl’s earlier work on Bayesian networks is helpful here, since it led to algorithms for more efficient search rules. These algorithms are nonetheless still very computationally intensive, and there has been very little work testing out their utility in the social sciences. Counterfactuals Pearl also argues that SCMs, and their implied probabilities, can be used to address seemingly intractable questions. Specifically, they can address unit-specific counterfactuals. Whereas interventions, and determining ATEs, can be performed by averaging across a group of cases, specific counterfactuals relate to an individual case. At first, counterfactuals seem unidentifiable. Think of a court case where there is an assertion that taking a drug caused a person’s death. There are two (potential) outcomes: \(Z_0\), the outcome when the person did not take the drug, i.e. \(X = 0\). \(Z_1\), the outcome when the person did take the drug, i.e. \(X = 1\). The person took the drug and died, so we know \(Z_1 = 1\) (\(1\) = death, \(0\) = no death). The defense would like to know \(P(Z_0 \mid X = 1, Y = 1)\). But this seems like nonesense. We want to know the probability of an event under one hypothetical world while conditioning on another world, the one we observed. The solution relies on establishing an SCM that explicitly includes error terms. Each of the \(U \in \{UX, UY, UZ\}\) is an individual-specific value. After fitting the model using the observed data, we can get these values for a specific person. We then alter the graph by setting the value of \(X\) or \(Y\) to the counterfactual value and solve for \(Z\) using the error term value identified by the full regression. In the most simplistic case, we are assuming that each person’s error term is determined exactly by the equations. Pearl’s texts also discuss working with stochastic errors to come up with bounds on possible counterfactuals. SCMs and ML Pearl (2018) makes the audacious claim that current machine learning models cannot ever assert causality because they cannot deal with interventions, let alone counterfactuals. A machine learning model takes a set of features \(V = \{v_1, v_2, \ldots, v_k\}\) and finds a function \(f_z\) mapping this set onto an outcome \(Z\). Using variations on statistical modeling, this amounts to modeling the joint distribution of all variables. However, using Pearl’s \(do(\cdot)\) operator, a joint distribution changes when we intervene on a variable. For example, if we are given a data set without knowing where it came from, we can fit a regression model using the joint distribution. Yet nothing about the join distribution tells us whether \(X\) is randomized or not. Causality requires knowing which conditional probabilities are invariant to changes in the structural model. ML is blind to this. ML as currently practiced throws a bunch of stuff into a blender and sees what comes out, akin to 20th century regression modeling that taught us to “control for everything.” This may not matter when we want to predict the presence of a dog, cat, or hot dog in a picture. It will matter if we want to: Tell policymakers whether or not to increase the minimum wage. Determine if admissions criteria at a university are racially biased. Find a defendant guilty in a criminal trial. Determine a counterfactual for an individual for whom existing data are not representative. ML models are akin to the underwear gnome problem: Features. \(\dots\) Prediction! The black box hides the answer we need if we want to develop effective rules that lead to socially desirable outcomes. Limitations of the Pearlian Weltanschauung At the same time, Pearl’s dismissal of non-SCM approaches to modeling (potential outcomes, ML) are based on finding specific cases where these approaches fail, but he does not give a sense as to how often they fail. Take, for example, our apparently complicated model: We can identify the canonical set of adjuster variables, which will be valid if any valid set exists. adjustmentSets(g, "X", "Z", type = "canonical") ## { S, T, U, V, W, Y } We see that we can in fact “control for everything”. lm(Z ~ ., data = g_tbl) %__% tidy ## # A tibble: 8 x 5 ## term estimate std.error statistic p.value ## ## 1 (Intercept) 0.0236 0.0322 0.732 4.64e- 1 ## 2 X 0.815 0.0314 26.0 4.54e-114 ## 3 Y 0.562 0.0332 16.9 1.39e- 56 ## 4 V 0.604 0.0349 17.3 7.55e- 59 ## 5 W 0.583 0.0332 17.5 4.11e- 60 ## 6 T -0.0466 0.0395 -1.18 2.38e- 1 ## 7 S -0.0106 0.0398 -0.266 7.90e- 1 ## 8 U 0.0692 0.0395 1.75 8.05e- 2 We didn’t do too bad. The problem, of course, is that there are SCMs that do not have all IVs or features as a proper adjustment set. How bad our conclusions are will depend on how well our representation of reality is. Indeed, reading Pearl’s (co-authored) introductory textbook Causality: a Primer, one can’t help but be struck by how many of the estimators look just like the types of formulas that Rubin and colleagues have developed using the potential frameworks approach. Is a complete re-orientation of applied statistics really going to result in different (and, presumably better) estimators? The jury is still out. Finally, not all SCMs are identified, especially when stepping away from the world of linearity. Reverse causation plagues observational studies of social behavior, and unless you are satisfied with declaring “congruence”, not even SCMs may save you. At best, given complicated nonlinear and nonrecursive systems of equations, checking the model-implied conditional probability will rule out some models, but certainly not all candidates.
          Data Engineer (Data Warehouse) - Bandwidth - Raleigh, NC      Cache   Translate Page      
Machine Learning & statistics experience. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Mon, 25 Feb 2019 20:43:17 GMT - View all Raleigh, NC jobs
          Senior Software Engineer, Network Engineering - Bandwidth - Raleigh, NC      Cache   Translate Page      
You've got experience with machine learning and anomaly detection. We need developers who learn quickly and are comfortable diving into the operational...
From Bandwidth - Tue, 19 Feb 2019 16:35:34 GMT - View all Raleigh, NC jobs
          Database Admin Intern - Bandwidth - Raleigh, NC      Cache   Translate Page      
Familiarity with Machine Learning & Statistical concepts. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Tue, 20 Nov 2018 22:34:57 GMT - View all Raleigh, NC jobs
          AI For Businesses      Cache   Translate Page      
Robot Arm And Hand
AI: you can’t seem to escape the talk around how computers and machine learning will take over all processes in business. And not participating in it, is like becoming a dinosaur. In fact you're probably using AI without even realising it and in this article we look at some of the types of AI and how they can benefit you in business.
          Technical Architect - AWS - CDW - Seattle, WA      Cache   Translate Page      
DevOps, Big Data, Machine Learning, Serverless computing etc. Solicit input/feedback from both internal and external customers to shape the service offering....
From CDW - Tue, 19 Feb 2019 18:49:04 GMT - View all Seattle, WA jobs
          Quality Assurance Engineer - Amazon.com Services, Inc. - Seattle, WA      Cache   Translate Page      
Do you want to use Amazon’s massive data sets and Machine Learning to do it? Our QAE will be able to understand software internals, debug problems using log...
From Amazon.com - Fri, 01 Feb 2019 10:10:42 GMT - View all Seattle, WA jobs
          Principal Associate Software Engineer - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), API Gateway, AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Kafka,...
From Capital One - Sat, 02 Feb 2019 15:41:05 GMT - View all Tysons Corner, VA jobs
          Advanced Threat Incident Response Lead - CGI Veterans Jobs - Fairfax, VA      Cache   Translate Page      
Experience with Machine Learning & Artificial Intelligence. It is an extraordinary time to be in business. Network analysis experience with Network sensors...
From CGI Veterans Jobs - Wed, 06 Mar 2019 01:25:21 GMT - View all Fairfax, VA jobs
          Technical Architect - AWS - CDW - Dallas, TX      Cache   Translate Page      
DevOps, Big Data, Machine Learning, Serverless computing etc. Solicit input/feedback from both internal and external customers to shape the service offering....
From CDW - Tue, 19 Feb 2019 18:49:04 GMT - View all Dallas, TX jobs
          Advanced Threat Incident Response Lead - CGI - Fairfax, VA      Cache   Translate Page      
Experience with Machine Learning & Artificial Intelligence. It is an extraordinary time to be in business. Network analysis experience with Network sensors...
From CGI - Tue, 05 Mar 2019 23:48:49 GMT - View all Fairfax, VA jobs
          SOF Network Engineer - Dynanet Corporation - Springfield, VA      Cache   Translate Page      
UNIX, Linux, or Cisco IOS. Machine learning, data mining, and knowledge discovery. Cisco CCNP Cloud, CCNP Routing and Switching....
From Indeed - Thu, 21 Feb 2019 22:02:49 GMT - View all Springfield, VA jobs
          Special Operations Network Engineer - Barbaricum - Springfield, VA      Cache   Translate Page      
UNIX, Linux, or Cisco IOS. Machine learning, data mining, and knowledge discovery. Cisco CCNP Cloud, CCNP Routing and Switching....
From Barbaricum - Wed, 17 Oct 2018 18:06:35 GMT - View all Springfield, VA jobs
          Technology Summer 2019 Intern - Machine Learning Developer - Time Warner - Atlanta, GA      Cache   Translate Page      
Turner Broadcasting System, Inc. Technology Summer 2019 Intern - Machine Learning Developer....
From Time Warner - Tue, 18 Dec 2018 21:08:34 GMT - View all Atlanta, GA jobs
          Machine Learning project      Cache   Translate Page      
I need to develop an electricity customer usage pattern from their meter reading (Budget: $30 - $250 USD, Jobs: Data Mining, Machine Learning)
          Transcription service Otter launches enterprise app for teams      Cache   Translate Page      
The Otter app uses AI and machine learning to render accurate transcripts that identify speakers, suggest keywords, and allow keyword searching.
          Do You Want an Intern Position at Autodesk      Cache   Translate Page      
At Autodesk, we view our interns as real world employees and start them doing research and work collaborating with our teams. Ultimately we want to develop the interns into employees and they want to return after they graduate. Many of our employees started as interns in all of our business divisions including Autodesk Research where our interns typically work on research projects such as design, robotics, AI, machine learning, generative design, additive manufacturing, future casting and story telling, and much...
          Foghorn Systems: Product Overview and Insight      Cache   Translate Page      
PRODUCT OVERVIEW: FogHorn’s software platform brings advanced analytics and machine learning to the on-premises edge environment, enabling a new class of applications for advanced monitoring and diagnostics, machine performance optimization, proactive maintenance and operational intelligence use cases.
          Sr Principal Cognitive Sftwr - Explainable AI      Cache   Translate Page      
Northrop Grumman Mission Systems in Beavercreek, Ohio sector is seeking a Sr Principal Cognitive Sftwr who will be an integral part of a Research, Technology Transition and Systems development Team that performs deep learning on problems ranging from Machine Translation, Automated Speech Recognition, Speech Synthesis, Image Processing, Cyber Solutions and Remote Sensing Applications. The selected applicant will have the opportunity to advance the state of the art for the intelligence production and analysis. The applicant will also have the opportunity to perform independent research and development. Conducts research in artificial intelligence (AI)/machine learning, and prototypes advanced machine learning and deep learning techniques to stretch the capability of autonomous systems research and development programs. Defines, develops, and delivers novel mathematical and statistical modeling and algorithm development to tackle the challenges of prediction, optimization, and classification.
          Data Science Developer      Cache   Translate Page      
NJ-Jersey City, Primary Skills: Python, R, Big Data, VLDB, Machine Learning & Statistical Analysis Description: Machine Learning & Data Science best practices Experience on VLDB, multi-structured, big data Statistical Analysis Required: Typically requires a minimum of 7-10+ years as a practitioner of machine learning in Financial Markets. Hands on Machine Learning developer in the Analytics Team for the digital t
          Stop Child Abuse Before it Happens with New Open Source Geospatial Machine Learning Tools      Cache   Translate Page      
Predict-Align-Prevent and Urban Spatial Analysis share an original open source geospatial machine learning framework for the prevention of child …
          Artificial Intelligence Specialist - IBM Watson - Accenture - Austin, TX      Cache   Translate Page      
IBM Cloud Platform, IBM Watson AI / Data / Vision / Speech/ Language, Machine Learning, Intelligent Automation, APIs....
From Accenture - Tue, 20 Nov 2018 20:31:24 GMT - View all Austin, TX jobs
          How Data Science Can Boost Customer Experience For Rapid Business Growth      Cache   Translate Page      

While content marketing may still remain the number one means to reach out to the customers on the internet, but in 2019 making use of data science is also the number one way to keep your customers around once you have caught their attention on the web.
Data science which is also synonymous to big data or machine learning helps businesses that are already using easy to use CRM software platforms to develop a kind of customer experience or CX that leave consumers of your offerings satisfied and looking for more.


          Implementation Engineer - Behavox - New York, NY      Cache   Translate Page      
Knowledge and application of statistical methods and Machine Learning algorithms. The IM team is owning an increasing number of deployment and integration tasks...
From Behavox - Mon, 10 Dec 2018 16:19:54 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
          Supervisor Test Engineering (6)- 69876 - Advanced Micro Devices, Inc. - Austin, TX      Cache   Translate Page      
Follow tech trends and understand impacts to AMD business. Machine learning experience a plus. What you do at AMD changes everything....
From Advanced Micro Devices, Inc. - Mon, 10 Dec 2018 19:32:25 GMT - View all Austin, TX jobs
          Computer Scientist, GS-1550-09/11 (DEU-DP) - US Department of the Interior - Middleton, WI      Cache   Translate Page      
Researches existing machine learning techniques, and applies appropriate methods towards prediction. For additional information on our internal telework policy,... $50,598 - $79,586 a year
From usajobs.gov - Mon, 04 Mar 2019 10:08:38 GMT - View all Middleton, WI jobs
          Vi ses på DevSum den 23-24 maj      Cache   Translate Page      
DevSum, sveriges häftigaste event för systemutvecklare - 23-24 maj 2019

Sveriges häftigaste konferens för systemutvecklare, DevSum är tillbaka och årets datum är 23-24 maj (förutom för dig som passar på att även boka in en heldagsworkshop med din favoritexpert den 22 maj, då blir det tre hela dagar med fullt fokus kod, arkitektur, säkerhet, AI, Machine Learning och mycket mera!

För mig personligen blir det första deltagandet och efter events som TechDays, Jfokus, Öredev mfl så vet jag ju att Svenska arrangörer är grymma på att leta upp de allra vassaste och mest intressanta talarna och framförallt är ju det alltid så skön stämning på dessa konferenser :)

Biljetterna och mer info hittar du såklart på www.devsum.se och de senaste uppdateringarna i sociala media från arrangörer och talare återfinns på Linked #DevSum19, Twitter #DevSum19, Facebook DevSum19 eller Instagram #DevSum19

Vi ses i Stockholm i maj!!
          Automation is Not Always the Answer      Cache   Translate Page      

One of the hottest topics in 2019 is automation through AI and machine learning. Katie Robbert tells us why automation isn’t always the answer

The post Automation is Not Always the Answer appeared first on Spin Sucks.

       

Related Stories

 

          Comparison between R and Python for recommendation systems      Cache   Translate Page      
I have the following files in R and Python: 1. UBCF with cosine similarity 2. IBCF with cosine similarity 3. UBCF with correlation 4. IBCF with correlation Please compare and conclude which method (using... (Budget: $30 - $250 USD, Jobs: Machine Learning, Python, R Programming Language, Statistical Analysis, Statistics)
          Computer Scientist, GS-1550-09/11 (DEU-DP) - US Department of the Interior - Middleton, WI      Cache   Translate Page      
Researches existing machine learning techniques, and applies appropriate methods towards prediction. For additional information on our internal telework policy,... $50,598 - $79,586 a year
From usajobs.gov - Mon, 04 Mar 2019 10:08:38 GMT - View all Middleton, WI jobs
          Dynamics CRM Developer - Concurrency - Brookfield, WI      Cache   Translate Page      
Combining two decades of CRM and ERP developments with innovative technology such as machine learning and automation, Dynamics 365 is helping businesses around...
From Concurrency - Wed, 30 Jan 2019 20:34:33 GMT - View all Brookfield, WI jobs
          Microsoft Dynamics 365 Engineer - Concurrency - Brookfield, WI      Cache   Translate Page      
Combining two decades of CRM and ERP developments with innovative technology such as machine learning and automation, Dynamics 365 is helping businesses around...
From Concurrency - Wed, 26 Dec 2018 20:30:31 GMT - View all Brookfield, WI jobs
          Comparison between R and Python for recommendation systems      Cache   Translate Page      
I have the following files in R and Python: 1. UBCF with cosine similarity 2. IBCF with cosine similarity 3. UBCF with correlation 4. IBCF with correlation Please compare and conclude which method (using... (Budget: $30 - $250 USD, Jobs: Machine Learning, Python, R Programming Language, Statistical Analysis, Statistics)
          How AI And Machine Learning Can Supercharge Your Social Media Marketing      Cache   Translate Page      

When most digital marketers hear the terms “artificial intelligence” or “machine learning,” they naturally think of Hollywood films about super-intelligent robots or computers powered by AI. As a result, they might assume AI couldn’t possibly...read more

The post How AI And Machine Learning Can Supercharge Your Social Media Marketing appeared first on Social Media Explorer.


          Senior Data Scientist - Cray - Seattle, WA      Cache   Translate Page      
Who is Cray? Our business is supercomputing. Working closely with other internal teams to integrate machine learning into existing product offerings using...
From Cray - Wed, 23 Jan 2019 08:18:39 GMT - View all Seattle, WA jobs
          [Freelancer] Machine learning project      Cache   Translate Page      
From Freelancer // Detection of Email Spoofing via Machine learning
          Supervisor Test Engineering (6)- 69876 - Advanced Micro Devices, Inc. - Austin, TX      Cache   Translate Page      
Follow tech trends and understand impacts to AMD business. Machine learning experience a plus. What you do at AMD changes everything....
From Advanced Micro Devices, Inc. - Mon, 10 Dec 2018 19:32:25 GMT - View all Austin, TX jobs
          Manager, Data Science - 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 - Fri, 30 Nov 2018 00:47:34 GMT - View all Boise, ID jobs
          Sr. Manager, Media Analytics - Micron - Folsom, CA      Cache   Translate Page      
Statistics, probability theory, heuristics and machine learning. This means conducting business with integrity, accountability, and partnership while supporting...
From Micron - Thu, 21 Feb 2019 18:56:44 GMT - View all Folsom, CA jobs
          Director Venture Capital - 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, 08 Feb 2019 00:52:01 GMT - View all Milpitas, CA jobs
          Support Tools Developer - Pure Storage - Lehi, UT      Cache   Translate Page      
The world is experiencing a revolution driven by next-generation technology like AI, machine learning, virtual reality, quantum computing, and self-driving cars...
From Pure Storage - Thu, 28 Feb 2019 00:41:54 GMT - View all Lehi, UT jobs
          Knowledge Manager - Pure Storage - Lehi, UT      Cache   Translate Page      
The world is experiencing a revolution driven by next-generation technology like AI, machine learning, virtual reality, quantum computing, and self-driving cars...
From Pure Storage - Fri, 22 Feb 2019 22:51:05 GMT - View all Lehi, UT jobs
          Technical Support Engineer I - 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, 23 Dec 2018 08:19:51 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 - Fri, 30 Nov 2018 08:21:49 GMT - View all Lehi, UT jobs
          Senior Data Scientist - Predictive Enterprise Group - Neudesic LLC - Philadelphia, PA      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 - Sat, 15 Dec 2018 21:58:12 GMT - View all Philadelphia, PA 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
          NSBE Intern- National Society of Black Engineers 2019 National Convention - Visa - Detroit, MI      Cache   Translate Page      
Strong research experiences and publication record in machine learning and/or data mining, advanced cryptography, systems security, blockchain and quantum...
From Visa - Wed, 09 Jan 2019 04:11:40 GMT - View all Detroit, MI jobs
          Sr. Security Analyst II - AbbVie - Lake County, IL      Cache   Translate Page      
Understanding of Machine Learning. Coordinate efforts among multiple business units during Response. Interpret and summarize technical information for...
From AbbVie - Thu, 10 Jan 2019 21:07:17 GMT - View all Lake County, IL jobs
          A Clear Definition of Machine Learning      Cache   Translate Page      

There’s a lot of buzz about machine learning in government today, given its potential to improve operations, cut costs and produce better program outcomes. But what exactly is it?

The post A Clear Definition of Machine Learning appeared first on GovLoop.


          All-Cloud vs. On-Premises vs. Hybrid: How 3 Businesses Chose the Right Infrastructure      Cache   Translate Page      
All-Cloud vs. On-Premises vs. Hybrid: How 3 Businesses Chose the Right Infrastructure juliet.vanwage… Mon, 03/11/2019 - 10:21

Carvana is not a run-of-the-mill ­ used-car dealership: Its unique method of selling and distributing vehicles nationwide is disrupting the industry. So it should come as no surprise that the company deploys cutting-edge technology and IT infrastructure.

Shoppers can peruse cars online, see a 360-degree view of vehicles through special camera technology, arrange financing and finalize the sale — all within a few minutes and without a salesperson’s help. Buyers can receive their cars at home or visit one of Carvana’s signature, fully automated car vending machines, where they plop in giant coins and watch the machine dispense their new ride.

To provide the best customer service possible and ensure 24/7 uptime, Imran Kazi, Carvana’s senior director of technology services, has taken a hybrid approach. He’s deployed a small footprint of Nutanix hyperconverged infrastructure at the company’s headquarters in Tempe, Ariz., for testing and ­development and at car inspection ­centers for local compute and storage needs. But the company’s website, ­applications and data are all hosted in the public cloud.

“We don’t want to reinvent the wheel by building a big data center infrastructure, then worrying about depreciating and failing hardware,” Kazi says. “We can focus on implementing and supporting the right software solutions and tools for our employees, which helps create better experiences for our customers.”

Carvana

Carvana's waiting room at their Tempe, Ariz., facility. Photo: Jim David.

More businesses and nonprofits are moving their infrastructure to the cloud, now that it’s proven to be secure, reliable and cost-effective. But while cloud service providers can improve efficiencies and provide benefits such as redundancy and the ability to scale up and down quickly, the cloud is not always the best choice or necessarily the least expensive.

In some cases, it’s less costly to retain applications and data in-house, particularly if organizations have legacy or custom applications that are not easily migrated to the cloud. And companies concerned about latency or that have to comply with data regulations may need to keep workloads on-premises.

Some organizations have gone all-in on the public cloud. Some are sticking with traditional on-premises infrastructure or a private cloud, while others are using a mix. Analysts say companies and nonprofits have to decide what’s right for their unique requirements.

“Some see hybrid as a step on the way to the cloud,” says Frank Della Rosa, IDC’s research director for SaaS and Cloud. “Workloads sitting on-premises can eventually be moved, but some ­businesses still deal with latency and bandwidth concerns or data sovereignty and privacy issues. Those factors ­determine whether a workload is suitable for the public cloud, on-premises or private cloud.”

SEE MORE: Get help sorting through the dizzying array of cloud and on-premises computing options.

1. Hybrid Infrastructure Delivers Flexibility for Carvana

When Kazi joined Carvana in 2016, he had the opportunity to build the company’s IT infrastructure from scratch. At the time, the online auto retailer had spun off from its former parent company, but its applications and data still resided in the former parent’s data centers.

Kazi migrated customer-facing and business applications, such as customer relationship management and enterprise resource planning software, across several cloud providers, including Microsoft Azure. He also adopted Software as a Service offerings: RingCentral for unified communications and Google G Suite for email and collaboration.

Carvana

Cars stacked in Carvana's vending machine at the Tempe, Ariz., facility. Photo: Jim David.

He chose the commercial cloud for three primary benefits: It saves money. It provides the company the agility it needs for growth. And instead of having to manage infrastructure in the production environment, the cloud frees up IT staff for more strategic needs.

Kazi also built a private cloud across two data centers using Nutanix’s HCI appliances, which combine servers, storage, networking and virtualization into a small-footprint appliance.

These unified systems, which run on Nutanix’s license-free AHV hypervisor, are more energy efficient, easier to ­manage and easier to scale than traditional hardware. If the environment reaches capacity, Kazi can purchase new appliances and quickly configure them with Nutanix’s management software.

“We can easily expand as we grow and have more projects,” he says.

Since Kazi’s arrival, Carvana has grown rapidly nationwide, expanding from 11 cities to 96. It runs about 350 virtual machines across 25 clusters of Nutanix appliances. Data scientists, ­analysts and developers use them as a test and staging area for new applications and algorithms. The company uses analytics to understand customer preferences, discover new markets to enter and drive logistical efficiencies.

Through the private cloud, employees can fully test out applications and algorithms before deploying them in production in the public cloud. Having full control of in-house infrastructure aids that effort.

“We can understand every nuance of the software and what resources it takes, and then we figure out the best cloud platform to put it into production,” Kazi says.

The company also installs a Nutanix appliance in each of its car inspection centers, where 360-degree photo ­technology takes data-intensive pictures of the cars. “We need the infrastructure locally to upload the photos to the cloud,” Kazi says.

Carvana car vending machine coin slot

At Carvana’s unique vending machines, buyers deposit special coins to retrieve the cars they purchase. Photo: Jim David.

2. The Cloud Saves BARBRI Time and Money

The BARBRI Group, a Dallas-based legal education company that provides online prep courses for bar exams, has saved hundreds of thousands of ­dollars annually by shutting down its data centers and moving fully to Microsoft Azure.

About three years ago, the company had two redundant data centers in colocation facilities, but the IT infrastructure was reaching its end of life. Faced with a massive IT investment, the company went with the cloud because it was more cost-effective and efficient, says IT Director Mark Kaplan.

One reason is the scalability that the cloud offers to a seasonal business like BARBRI, which has its busy seasons in summer and winter. Instead of internal data centers running full-throttle year-round, Azure allows the company to spin up more resources during the demanding months and scale down during slow periods, Kaplan says.

It’s a huge financial savings,” he says. “We didn’t have to do a huge capital outlay, and then we ended up finding out that it’s just easier to manage.”

Cloud

Source: RightScale, “2018 State of the Cloud Report,” February 2018

BARBRI migrated to the cloud gradually. In 2016, Kaplan shut down one data center and operated half on-prem and half in the cloud. The company migrated to Azure’s Database as a Service offering, then it moved 250 applications.

“It allowed us to get used to working in the cloud, and if it failed, we could cut back to on-premises,” he explains.

The cloud effort was a success, however. So, in January 2017, Kaplan shut down the remaining data center and went all-cloud. The company uses Azure Backup for its virtual machines and changed from an expensive multiprotocol label-switching network to a ­software-defined WAN using Cisco Meraki switches. Microsoft manages the infrastructure and databases, while the BARBRI IT staff manages its own ­software and handles OS upgrades. The freed-up time allows the IT staff to pursue new projects, such as using machine learning to build online courses that adapt to students’ needs in real time.

Modern-Workforce_the-office.jpg

3. Detroit Symphony Orchestra Sees the Advantages of HCI

At the Detroit Symphony Orchestra, Jody Harper knew the organization’s three aging servers were near the end of life, but as a nonprofit with a tight budget, it hoped to prolong their life span as much as possible. In late 2017, however, the servers crashed, knocking out the symphony’s website and preventing its 75 full-time employees from working for a full day.

Harper got the 8-year-old servers running again, but it was just a temporary fix. He needed a long-term solution — and fast.

“If no one can buy tickets from our website, that’s money lost,” says Harper, senior director of technology and infrastructure.

Fortunately, he had done his homework and researched his options. Moving to the cloud was not realistic because it would require too much time and money to redesign the custom integration that exists between the symphony’s two most critical applications — its website and CRM software — for the cloud.

Instead, Harper wanted hyperconverged equipment. In the ensuing four weeks, he purchased and deployed three Scale Computing hyperconverged appliances because of their affordability and ease of use.

“For a nonprofit, cost is, hands-down, a major factor,” he says. “It’s also simple and easy. I can spin up a new server in 10 clicks.”

Today, 25 virtual servers run on the Scale equipment, including financial and event management software, databases and file and print servers.

A hybrid cloud approach is still a future option, however. To improve disaster recovery, Harper is considering using the public cloud as a backup site. In the meantime, he recently purchased a fourth Scale appliance to replicate data in a separate location.

“It’s a reliable product. It’s made our lives easier, and now we have a fourth one for disaster purposes,” he says.


          Roasted corn — and machine learning in a food truck — at 23rd and Union’s El Costeño      Cache   Translate Page      
Moises Santos is a 24-year-old programmer, food truck designer, and immigrant from Oaxaca, Mexico. His food truck holds down what seems like prime territory — the pot purchasing and stoner friendly parking lot at the Central District’s Uncle Ike’s. The … Continue reading
          Google rolls out AR effect tools for YouTube Stories      Cache   Translate Page      

MUMBAI: With the overwhelming popularity of Stories feature on social media platforms, the companies are leaving no stones unturned to make it more attractive. Now, Google is rolling out support for its advanced Augmented Reality (AR) effect tools for YouTube Stories while its rival Facebook and Instagram already support AR filters for Stories on their platforms.

The new feature will allow users to add animated masks, glasses, 3D hats and more such objects to their selfies. "To make all this possible, we employ machine learning (ML) to infer approximate 3D surface geometry to enable visual effects and ML pipeline for Selfie AR," Google Artificial Intelligence research engineers Artsiom Ablavatski and Ivan Grishchenko wrote in a blog post on Saturday.

"That way we can grow our dataset to increasingly challenging cases, such as grimaces, oblique angle and occlusions. Dataset augmentation techniques also expanded the available ground truth data, developing model resilience to artefacts like camera imperfections or extreme lighting conditions," the post added.

The company has also claimed that it will use improved "anchoring" process with the new AR effects to make them look more real and responsive. According to the post, it uses a unique set of technologies "that can track the highly dynamic surface geometry across every smile, frown or smirk."

However, YouTube Stories is not available to every user yet as the tech giant launched the feature last year but only to creators with more than 10k subscribers.

http://www.indiantelevision.com/sites/default/files/styles/300x300/public/images/tv-images/2019/03/12/youtube.jpg?itok=eYDItqfz

          Primer: Kubeflow Streamlines Machine Learning with Kubernetes      Cache   Translate Page      

This article is a post in a series on bringing continuous integration and deployment (CI/CD) practices to machine learning. Check back to The New Stack for future installments. Kubeflow was created to make it easier to develop, deploy and manage machine learning applications. It’s a composable, scalable, portable machine learning stack based on Kubernetes that […]

The post Primer: Kubeflow Streamlines Machine Learning with Kubernetes appeared first on The New Stack.


          Machine Learning Team Leader - DayTwo Inc - Tocantins      Cache   Translate Page      
DayTwo is looking for an exceptional and motivated team leader with strong applied and research machine learning abilities to join our team. DayTwo is heavily...
De DayTwo Inc - Sun, 24 Feb 2019 16:13:38 GMT - Visualizar todas as empregos: Tocantins
          Artificial intelligence cuts lung cancer screening false positives      Cache   Translate Page      
(University of Pittsburgh) Right now, 96 percent of people who screen positive for lung cancer don't actually have a malignant growth. Machine learning can rule out cancer in a third of them, saving time, money and anxiety.
          Mathematical Strategies for Design Optimization of Multiphase Materials      Cache   Translate Page      
This work addresses various mathematical solution strategies adapted for design optimization of multiphase materials. The goal is to improve the structural performance by optimizing the distribution of multiple phases that constitute the material. Examples include the optimization of multiphase materials and composites with spatially varying fiber paths using a finite element analysis scheme. In the first application, the phase distribution of a two-phase material is optimized to improve the structural performance. A radial basis function (RBF) based machine learning algorithm is utilized to perform a computationally efficient design optimization and it is found to provide equivalent results with the physical model. The second application concentrates on the optimization of spatially varying fiber paths of a composite material. The fiber paths are described by the Non-Uniform Rational Bezier (B)-Spline Surface (NURBS) using a bidirectional control point representation including 25 parameters. The optimum fiber path is obtained for various loading configurations by optimizing the NURBS parameters that control the overall distribution of fibers. Next, a direct sensitivity analysis is conducted to choose the critical set of parameters from the design point to improve the computational time efficiency. The optimized fiber path obtained with the reduced number of NURBS parameters is found to provide similar structural properties compared to the optimized fiber path that is modeled with a full NURBS representation with 25 parameters.
          New Chrome AI custom app Tune hides toxic comments      Cache   Translate Page      
A new open source Chrome app called Tune has been developed by Alphabet/Google subsidiary Jigsaw. It hides and moderates toxic social media comments with artificial intelligence using machine learning techniques.
          Senior Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sat, 05 Jan 2019 10:09:26 GMT - View all New York, NY jobs
          Machine Learning Engineer - Temboo - New York, NY      Cache   Translate Page      
You will lead the design, prototyping and productization of machine learning-based features, and take responsibility for introducing other Temboo developers to...
From Temboo - Thu, 29 Nov 2018 10:09:26 GMT - View all New York, NY jobs
          Engineering Manager - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sun, 28 Oct 2018 10:09:38 GMT - View all New York, NY jobs
          Frontend Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role building web-based features on top of transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:31 GMT - View all New York, NY jobs
          Embedded Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:30 GMT - View all New York, NY jobs
          Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Fri, 28 Sep 2018 10:09:40 GMT - View all New York, NY jobs
          Jigsaw’s new Chrome extension will ‘Tune’ out toxic YouTube, Twitter, Facebook, Disqus comments      Cache   Translate Page      

Jigsaw is an Alphabet incubator tasked with using technology to tackle global security challenges. Notable projects include protections against DDoS attacks and DNS manipulation. Its latest is a Chrome extension called Tune to filter out toxic online comments with machine learning.

more…

The post Jigsaw’s new Chrome extension will ‘Tune’ out toxic YouTube, Twitter, Facebook, Disqus comments appeared first on 9to5Google.


          BUSINESS INTELLIGENCE ANALYST - TransUnion - Portland, OR      Cache   Translate Page      
Identify and explore a creative project that utilizes your strengths in machine learning, business process development, global statistical analysis, or data...
From TransUnion - Wed, 06 Feb 2019 04:09:51 GMT - View all Portland, OR jobs
          SR. RESEARCH AND CONSULTING 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 - Tue, 29 Jan 2019 04:50:07 GMT - View all Chicago, IL jobs
          SR. RESEARCH AND CONSULTING MANAGER - 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 - Tue, 27 Nov 2018 10:04:56 GMT - View all Chicago, IL jobs
          Speech to Text & Voice Chatbots on IVR      Cache   Translate Page      
voice bots will answer the call and will ask further queries to get information from customer. The conversation will be converted into text directly, (Budget: ₹75000 - ₹150000 INR, Jobs: Machine Learning)
          Business Development Manager - Machine Learning Field Leader - Amazon Web Services, Inc. - Seattle, WA      Cache   Translate Page      
Do you love building new businesses? AWS customers are looking for ways to change their business models and solve complex business challenges with artificial...
From Amazon.com - Tue, 08 Jan 2019 09:38:28 GMT - View all Seattle, WA jobs
          Sr. Data Scientist - Worldwide Public Sector Team - Amazon Web Services, Inc. - Herndon, VA      Cache   Translate Page      
Travel for face-to-face customer engagements, internal conferences, and industry events. Machine learning has been strategic to Amazon from the early years....
From Amazon.com - Tue, 08 Jan 2019 09:37:16 GMT - View all Herndon, VA jobs
          Speech to Text & Voice Chatbots on IVR      Cache   Translate Page      
voice bots will answer the call and will ask further queries to get information from customer. The conversation will be converted into text directly, (Budget: ₹75000 - ₹150000 INR, Jobs: Machine Learning)
          Optimizing the depth and the direction of prospective planning using information values      Cache   Translate Page      
by Can Eren Sezener, Amir Dezfouli, Mehdi Keramati Evaluating the future consequences of actions is achievable by simulating a mental search tree into the future. Expanding deep trees, however, is computationally taxing. Therefore, machines and humans use a plan-until-habit scheme that simulates the environment up to a limited depth and then exploits habitual values as … Continua la lettura di Optimizing the depth and the direction of prospective planning using information values
          Individual prognosis at diagnosis in nonmetastatic prostate cancer: Development and external validation of the PREDICT Prostate multivariable model      Cache   Translate Page      
by David R. Thurtle, David C. Greenberg, Lui S. Lee, Hong H. Huang, Paul D. Pharoah, Vincent J. Gnanapragasam Background Prognostic stratification is the cornerstone of management in nonmetastatic prostate cancer (PCa). However, existing prognostic models are inadequate—often using treatment outcomes rather than survival, stratifying by broad heterogeneous groups and using heavily treated cohorts. To … Continua la lettura di Individual prognosis at diagnosis in nonmetastatic prostate cancer: Development and external validation of the PREDICT <i>Prostate</i> multivariable model
          Age distribution, trends, and forecasts of under-5 mortality in 31 sub-Saharan African countries: A modeling study      Cache   Translate Page      
by Iván Mejía-Guevara, Wenyun Zuo, Eran Bendavid, Nan Li, Shripad Tuljapurkar Background Despite the sharp decline in global under-5 deaths since 1990, uneven progress has been achieved across and within countries. In sub-Saharan Africa (SSA), the Millennium Development Goals (MDGs) for child mortality were met only by a few countries. Valid concerns exist as to … Continua la lettura di Age distribution, trends, and forecasts of under-5 mortality in 31 sub-Saharan African countries: A modeling study
          TensorFlow.js 1.0      Cache   Translate Page      

Rilasciato TensorFlow.js 1.0, prima major release della libreria JavaScript che consente di implementare modelli di machine learning da browser Web.

Leggi TensorFlow.js 1.0


          Machine Learning Team Leader - DayTwo Inc - Tocantins      Cache   Translate Page      
DayTwo is looking for an exceptional and motivated team leader with strong applied and research machine learning abilities to join our team. DayTwo is heavily...
De DayTwo Inc - Sun, 24 Feb 2019 16:13:38 GMT - Visualizar todas as empregos: Tocantins
          Machine Learning - Accenture - Bengaluru, Karnataka      Cache   Translate Page      
Accenture Technology powers our clients’ businesses with innovative technologies—established and emerging—changing the way their people and customers experience...
From Accenture - Tue, 12 Mar 2019 15:46:15 GMT - View all Bengaluru, Karnataka jobs
          Technical Architect - AWS - CDW - Seattle, WA      Cache   Translate Page      
DevOps, Big Data, Machine Learning, Serverless computing etc. Solicit input/feedback from both internal and external customers to shape the service offering....
From CDW - Tue, 19 Feb 2019 18:49:04 GMT - View all Seattle, WA jobs
          Quality Assurance Engineer - Amazon.com Services, Inc. - Seattle, WA      Cache   Translate Page      
Do you want to use Amazon’s massive data sets and Machine Learning to do it? Our QAE will be able to understand software internals, debug problems using log...
From Amazon.com - Fri, 01 Feb 2019 10:10:42 GMT - View all Seattle, WA jobs
          Principal Associate Software Engineer - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), API Gateway, AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Kafka,...
From Capital One - Sat, 02 Feb 2019 15:41:05 GMT - View all Tysons Corner, VA jobs
          Technical Architect - AWS - CDW - Dallas, TX      Cache   Translate Page      
DevOps, Big Data, Machine Learning, Serverless computing etc. Solicit input/feedback from both internal and external customers to shape the service offering....
From CDW - Tue, 19 Feb 2019 18:49:04 GMT - View all Dallas, TX jobs
          Senior AI Solutions Architect - Industrial - Petuum - Sunnyvale, CA      Cache   Translate Page      
Machine learning or IIoT preferred. PaaS, SaaS, IaaS and business intelligence/analytics implementation experience are a plus....
From Petuum - Sun, 06 Jan 2019 08:08:10 GMT - View all Sunnyvale, CA jobs
          Legal Tool Among Winners Of Knight Foundation ‘AI and the News’ Challenge      Cache   Translate Page      
The Knight Foundation, an organization devoted to promoting excellence in journalism, today announced the seven winners of an open challenge to use artificial intelligence to empower journalism and reimagine news, and among them is a legal technology company. Legal Robot, a San Francisco company whose product uses machine learning to analyze contracts, won for its... Continue Reading
          How Has Machine Learning And AI Influenced Game Design?      Cache   Translate Page      
How has machine learning and AI influenced game design? This question was originally answered on Quora by Travis Addair.
          Machine Learning Model      Cache   Translate Page      
Need to build intelligent models which can predict form fields in the editable PDF form fields Will give you more details . TensorFlow or any other ML technique can be used (Budget: $2 - $8 USD, Jobs: Artificial Intelligence, Machine Learning, Python, Tensorflow)
          Machine learning Project      Cache   Translate Page      
Need to build intelligent models which can predict form fields in the editable PDF form fields Will give you more details . TensorFlow or any other ML technique can be used (Budget: $2 - $8 USD, Jobs: Algorithm, Machine Learning, Python, Software Architecture)
          DevOps Engineer - Vehicle Security Team - Cylance, Inc. - Texas      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, 12 Dec 2018 19:28:33 GMT - View all Texas jobs
          Director, Security Research - DigitalShield - White Plains, NY      Cache   Translate Page      
By successfully applying machine learning and artificial. Redefined the threat intelligence market, garnered acclaim from industry....
From DigitalShield - Fri, 22 Feb 2019 09:05:01 GMT - View all White Plains, NY jobs
          Consultant - DigitalShield - White Plains, NY      Cache   Translate Page      
By successfully applying machine learning and artificial. Detailing technical issues identified and their associated business....
From DigitalShield - Fri, 22 Feb 2019 09:05:01 GMT - View all White Plains, NY 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. - Thu, 03 Jan 2019 07:27:34 GMT - View all North Carolina jobs
          Technical Account Manager - 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, 25 Feb 2019 19:27:39 GMT - View all Irvine, CA 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. - Sat, 12 Jan 2019 14:06:39 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. - Sun, 06 Jan 2019 07:27:26 GMT - View all Irvine, CA jobs
          If you did not already know      Cache   Translate Page      
Generative Topographic Map (GTM) Generative topographic map (GTM) is a machine learning method that is a probabilistic counterpart of the …

Continue reading


          Software Engineer - Machine Learning - Convoy - Seattle, WA      Cache   Translate Page      
Today, we use machine learning to figure out freight prices, shipment relevance for carriers, auction bidding strategy, and other internal processes....
From Convoy - Wed, 16 Jan 2019 10:12:47 GMT - View all Seattle, WA jobs
          Operations Research Scientist, Data Science - Convoy - Seattle, WA      Cache   Translate Page      
Today, we use machine learning and economic analysis to figure out freight prices, shipment relevance for carriers, auction bidding strategy, and other internal...
From Convoy - Fri, 11 Jan 2019 04:12:37 GMT - View all Seattle, WA jobs
          Senior Data Engineer, Data Science - Convoy - Seattle, WA      Cache   Translate Page      
Data engineering, database engineering, business intelligence or business analytics. Today, we use machine learning to figure out freight prices, shipment...
From Convoy - Tue, 27 Nov 2018 22:12:32 GMT - View all Seattle, WA jobs
          Senior Applied Scientist - Convoy - Seattle, WA      Cache   Translate Page      
Enhancing our machine learning infrastructure. Today, we use machine learning to figure out freight prices, shipment relevance for carriers, auction bidding...
From Convoy - Mon, 22 Oct 2018 22:12:42 GMT - View all Seattle, WA jobs
          Sr. Director/VP, Marketing Sciences - Hypothesis Group - Los Angeles, CA      Cache   Translate Page      
Understanding of machine learning techniques and algorithms. Active support in research design for new business....
From Hypothesis Group - Fri, 11 Jan 2019 22:04:38 GMT - View all Los Angeles, CA jobs
          Senior Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sat, 05 Jan 2019 10:09:26 GMT - View all New York, NY jobs
          Machine Learning Engineer - Temboo - New York, NY      Cache   Translate Page      
You will lead the design, prototyping and productization of machine learning-based features, and take responsibility for introducing other Temboo developers to...
From Temboo - Thu, 29 Nov 2018 10:09:26 GMT - View all New York, NY jobs
          Engineering Manager - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sun, 28 Oct 2018 10:09:38 GMT - View all New York, NY jobs
          Frontend Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role building web-based features on top of transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:31 GMT - View all New York, NY jobs
          Embedded Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:30 GMT - View all New York, NY jobs
          Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Fri, 28 Sep 2018 10:09:40 GMT - View all New York, NY jobs
          Data Scientist Lead - Schneider National - Green Bay, WI      Cache   Translate Page      
Experience with machine learning software (e.g., R, Python, SPSS, SAS), data access/manipulation (e.g., SQL, pandas, dplyr) and NoSQL databases (e.g., MongoDB,...
From Schneider National - Thu, 03 Jan 2019 06:22:26 GMT - View all Green Bay, WI jobs
          Data Scientist Lead - Schneider - Green Bay, WI      Cache   Translate Page      
Experience with machine learning software (e.g., R, Python, SPSS, SAS), data access/manipulation (e.g., SQL, pandas, dplyr) and NoSQL databases (e.g., MongoDB,...
From Schneider - Wed, 02 Jan 2019 23:36:22 GMT - View all Green Bay, WI jobs
          NOC Service Desk Analyst II - Apptio - Morrisville, NC      Cache   Translate Page      
Skype for Business, WebEx). Datacenter and server hardware (Red Hat Linux). Apptio's software uses machine learning to translate technology costs and...
From Apptio - Fri, 22 Feb 2019 02:32:52 GMT - View all Morrisville, NC jobs
          Lead Software Engineer - Java - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Streaming....
From Capital One - Thu, 28 Feb 2019 18:12:30 GMT - View all Tysons Corner, VA jobs
          Principal Associate Software Engineer - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), API Gateway, AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Kafka,...
From Capital One - Sat, 02 Feb 2019 15:41:05 GMT - View all Tysons Corner, VA jobs
          AI Platform Software Architect - DISCO - Austin, TX      Cache   Translate Page      
Should have prior experience designing, implementing, and operating scalable distributed systems for Big Data, Machine Learning or Machine Learning Problems....
From Disco - Tue, 12 Mar 2019 16:09:59 GMT - View all Austin, TX 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 - Thu, 20 Dec 2018 07:48:32 GMT - View all Providence, RI jobs
          Bayesian Machine Learning in Python: A/B Testing      Cache   Translate Page      
Bayesian Machine Learning in Python: A/B Testing#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000
Bayesian Machine Learning in Python: A/B Testing
.MP4 | Video: 1280x720, 30 fps(r) | Audio: AAC, 48000 Hz, 2ch | 853 MB
Duration: 5.5 hours | Genre: eLearning Video | Language: English

Data Science, Machine Learning, and Data Analytics Techniques for Marketing, Digital Media, Online Advertising, and More.

Use adaptive algorithms to improve A/B testing performance

Apply Bayesian methods to A/B testing


          CorelDRAW vuelve al Mac por la puerta grande con su nueva suite completamente rediseñada para macOS, ya disponible en la Mac App Store       Cache   Translate Page      

CorelDRAW vuelve al Mac por la puerta grande con su nueva suite completamente rediseñada para macOS, ya disponible en la Mac App Store #source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

Uno de las grandes marcas del diseño gráfico, CorelDRAW ha anunciado hoy el lanzamiento de su nueva suite CorelDRAW Graphics Suite 2019 con importantes novedades relacionadas con el mundo Mac: una versión nativa completamente rediseñada desarrollada para macOS Mojave aprovechando al máximo las características del sistema operativo.

Según nos ha contado Klaus Vossen, Director Jefe de Productos en CorelDRAW: "nuestros usuarios nos pedían de forma recurrente que lanzásemos una nueva versión nativa aprovechando las posibilidades de los últimos Mac". Para ello, la nueva versión para Mac es exactamente idéntica a la de Windows - no deja nada por el camino - y tiene las mismas características y funcionalidades.

Corel Draw Event Applesfera

Los piezas clave que definen a esta nueva suite incluyen además de la famosísima CorelDRAW de diseño vectorial, edición de fotografías, alto rendimiento y herramientas de dibujo basadas por primera vez en inteligencia artificial. En este sentido, el software utilizará el Machine Learning para reconocer trazos realizados con lápices digitales y asistirnos en los diseños, aprendiendo de nosotros cuanto más lo usamos.

La suite completa 2019 llega al Mac en buena forma

Coreldraw Graphics Suite 2019 For Mac Right

Esta nueva versión para Mac refleja en cierta forma aspectos prácticos de macOS que esperan los usuarios de Mac: interfaz adaptada a nuestro sistema operativo, compatibilidad con el modo oscuro y la Touch Bar del MacBook Pro - que cambiará dinámicamente dependiendo de la herramienta con la que estemos trabajando.

La suite se compone de cinco aplicaciones:

  • CorelDRAW, la conocida herramienta de diseño vectorial, ilustración y diseño de páginas.
  • Corel Photo-Paint, para edición de fotografías.
  • Corel Font Manager, para indexar y organizar bibliotecas De Fuentes.
  • AfterShot 3 HDR, para procesar imágenes RAW.

Las principales novedades de esta nueva versión de CoreDRAW son las ventanas acoplables de objetos, efectos no destructivos, diseño que se alinea a los píxeles, nuevas plantillas y un nuevo aspecto renovado y modernizado que potencia el rendimiento general y busca la creatividad de los usuarios.

Coreldraw 2019 For Mac Symmetry Mode Es

Para sorpresa de todos, Corel también ha tenido hoy un One More Thing: una nueva versión web a la que podemos acceder mediante navegador, sea cual sea la plataforma y el sistema operativo. Esta nueva opción está incluida en los planes de precio de la suite completa.

En Applesfera estamos preparando un análisis a fondo del nuevo Creative Suite 2019, pero la nueva suite está ya disponible en la Mac App Store mediante dos opciones: licencia perpetua y licencia de suscripción. Corel nos ha contado que no querían forzar las opciones de sus usuarios a la hora de comprar el producto.

recommendation.header

Gasolina, diésel, híbrido… Cómo acertar con la propulsión

Pixelmator 3.1, más potente y rápido que nunca gracias al nuevo Mac Pro

Hay luz más allá de Adobe: así esta el panorama de precios de las apps de diseño en OS X

-
the.news CorelDRAW vuelve al Mac por la puerta grande con su nueva suite completamente rediseñada para macOS, ya disponible en la Mac App Store originally.published.in por Pedro Aznar .


          Machine Learning Engineering Intern - EY - Palo Alto, CA      Cache   Translate Page      
AI Labs in Palo Alto, CA, under the leadership of the Global AI Leader, Nigel Duffy, is in the process of revolutionizing EY and the the auditing industry with...
From EY - Thu, 07 Mar 2019 01:39:45 GMT - View all Palo Alto, CA jobs
          Web/Server Software: Kubeflow on OpenShift, HTTP, Rspamd and Splunk (Proprietary)      Cache   Translate Page      
  • Kubeflow on OpenShift

    Kubeflow is an open source project that provides Machine Learning (ML) resources on Kubernetes clusters. Kubernetes is evolving to be the hybrid solution for deploying complex workloads on private and public clouds. A fast growing use case is using Kubernetes as the deployment platform of choice for machine learning.

    Infrastructure engineers will often spend time modifying deployments before a single model can be tested. These deployments are often bound to the clusters they have been deployed to, thus moving a model from a laptop to a cloud cluster is difficult without significant re-architecture.

  • Daniel Stenberg: Looking for the Refresh header

    The other day someone filed a bug on curl that we don’t support redirects with the Refresh header. This took me down a rabbit hole of Refresh header research and I’ve returned to share with you what I learned down there.

  • Rspamd 1.9.0 has been released
  • 12 Splunk User and Role Administration Examples for both CLI and Web

    Splunk supports three types of authentication: Native Authentication, LDAP and Scripted Authentication API.

    For most part, Native Authentication is referred as Splunk authentication, which takes high priority over any external authentication.


          Foghorn Systems: Product Overview and Insight      Cache   Translate Page      
PRODUCT OVERVIEW: FogHorn’s software platform brings advanced analytics and machine learning to the on-premises edge environment, enabling a new class of applications for advanced monitoring and diagnostics, machine performance optimization, proactive maintenance and operational intelligence use cases.
          High German release AI engine, a high German two strategic landing – Sohu Technology-douke      Cache   Translate Page      
High German AI engine released "a high German two center" strategic landing technology Sohu – Sohu technology paper Cui Peng September 19th High German map released in Beijing High German map AI engine, this engine is based on large data capacity and machine learning ability as the basis, for different environments and needs, provided travel […]
          Data Scientist - Oliver Wyman - New York, NY      Cache   Translate Page      
Demonstrate solid and battle-tested understanding of the standard canon of machine learning practices, including but not limited to:....
From Marsh & McLennan Companies - Sat, 05 Jan 2019 15:04:15 GMT - View all New York, NY jobs
          PL SQL Practical Live Online Training       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 ...
          Best Project Oriented Online Training On Power BI       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 ...
          Supervisor Test Engineering (6)- 69876 - Advanced Micro Devices, Inc. - Austin, TX      Cache   Translate Page      
Follow tech trends and understand impacts to AMD business. Machine learning experience a plus. What you do at AMD changes everything....
From Advanced Micro Devices, Inc. - Mon, 10 Dec 2018 19:32:25 GMT - View all Austin, TX jobs
          I'd love to...      Cache   Translate Page      
It's been a while since I've found the time to write a blog post. I'm writing for Eli Lilly now, and you can check out some of those posts here --->  https://lillypad.lilly.com/?auth=76 I find a lot of the content I'm sharing over there to be stuff that I would also share on here, so no sense in duplicating!

We decorated for Christmas the day after Thanksgiving, marking the start of my most favorite holiday and time of the year. Christmas music is being piped throughout the house (and car and headphones) and reflections of the past year have inevitably begun. We had a tough go for the second half of this year, and we're sort of still in it. If you recall, Drew lost a significant amount of his lung function over the summer before we identified the culprit - a fungal infection. We treated him with anti-fungal medication and he thankfully improved until there was a second unexpected drop near the start of the school year. He had gotten back up to 91, and then in September dropped back down to 78. We had discussions about what might be causing it and what we should do to treat him, and decided to temporarily stay the course on the anti-fungal medication until he reaches and maintains a baseline on it, giving us confidence that it is both working and that the infection is under control. You may recall (or maybe not) that he had a fungal infection last year around this time. We started the anti-fungals for 3mo and he improved, so we stopped the treatment, and then by June he had lost 30% of his lung function, maybe not so mysteriously after all. I suggested that perhaps we hadn't had the infection under control as we had originally thought and, like fungus does, it slowly crept back wreaking silent havoc. I want to make sure that we are confident that things are under control this time before we change course, as a newer article suggests that fungus can become quite resistant if treated, if the medications used to treat it aren't used properly. The options seem to be, per this article anyway, treat the fungus and increase its adaptive skills, or not treat it and allow the pathogen to settle in the lungs. Not treating wasn't an option for us because of the impact that it was having on his lung function.

It's complicated, this disease. We seem to have the bacterial load in his lungs under control. Most research shows that bacterial exacerbations are a leading cause of lung function decline and lung damage in CF. But once we finally got the achromobacter under control with years of treatment on inhaled antibiotics and steroids, we seem to have traded it for a fungal infection. Did we cause the fungal infection? Perhaps we did, there's not really a good way to say. Is it better to have a bacterial infection or a fungal infection? I would probably argue that a fungal infection is *better* given that these is little research that shows the impact of a fungal infection on the progression of disease. I do not know if there is evidence to support the contrary, or just lack of research on this altogether. Either way, I'm interested to learn more and hope that the CFF will continue to study this.

He's got a cold now, coughing in his sleep and when he's running around and playing. We've added extra treatments which has him crankier than ever, but it's necessary. We've been going in to clinic for PFT's every two weeks and his numbers are remaining pretty consistent - 78, 81, 82 - but I'm not so sure how things will look with this new cold, perhaps something he picked up when we were in clinic for one of those appointments. Despite their best infection control practices - recently even declaring that the spread of infection among patients in our clinic had come to a halt with new infection control practices, which is great news - going into the hospital remains one of the most dangerous places for Drew to be. There are lots of sick people coming here for care, and even though we wear a mask and don't touch anything, he always seems to catch something when we have to come here. We should be able to use home spirometers to monitor our lung function. We should be able to track our weight from home, and other symptoms, and communicate what we learn with our care team, eliminating the need for unnecessary visits, saving everyone time and money, and perhaps even improving health. Machine learning can enable this, and should. While our center was using the Orchestra platform (which is no longer), we did start to see a longitudinal view of patients health shared with the care team. We did improve inter-visit communication, and intervention at more appropriate times rather than just when we happened to have a visit scheduled. It didn't reduce the number of times that we *needed* to come into clinic but it could have. I highlight the word *needed* as this is an evidence based medicine metric, a guideline put forth by the CFF for all patients, and embraced by all clinicians, regardless of whether its the right thing for the patient. The care teams aren't interested in reducing clinic visits below the required 4x a year. Or maybe they are interested but just can't becasue the CFF uses this as an accreditiation metric, requiring them to do this or find a way to improve rather than working to understand, from patients, why they aren't coming in 4x/yr and how we might work together to optimize care and outcomes according to the patient priorities. Hopefully our learning network will fix this. I digress.

I hope he's well for the holidays. I got this crazy idea to take my family to NYC to see some cousins the week before Christmas. What crazy person wouldn't want to drive 24hrs over 3 days to spend a night in a matchbox sized hotel room to see family and New York City at Christmastime?! I'm sure traffic will be delightful. At least we can stream Christmas music in the car!

I hope he feels well enough to open gifts with excitement and delight on Christmas morning, and that we don't have to pull him away from new toys to do extra treatments. I'd love to, for just one day, forget all of the medical stuff. I'd love to wake up and not have a schedule. I'd love to go out to dinner and not worry about hand sanitizer and enzymes. I'd love for him to run around outside, maybe in the snow, and not come back inside having a coughing fit, and rather than breathing treatments, have hot chocolate while all of the clothes defrost into a puddle in my foyer.

I've declined antibiotics for him since September because I don't think he needs them. Maybe I'm wrong, but I'm trusting my gut this time. We will go there if we need to, but for right now, we will do our treatment and take our medicine and listen to our Christmas music and enjoy this holiday as much as we can.
          Let's talk about evidence      Cache   Translate Page      
"We don't seek the painful experiences that hue our identity, but we seek our identity in the wake of painful experiences."

I don't know where this quote came from, only that I find so much truth in it. If whoever wrote it happens to read this, raise your hand so I can learn more from you.

Cystic Fibrosis sucks so bad. You can't usually see it, yet it forces you to act - restraining your kid for necessary blood work to make sure the medication we're taking to save his lungs isn't killing his liver. Childhood playtime after school is stolen from us as we sit for hours doing breathing treatments and airway clearance instead of going to the park without an agenda. A hug allows me to literally feel a mucus plug trapped deep in his airways, something I can loosen with a few aggressive beats on his chest and I do it without warning as he would deny my offer to help because of the pain it causes him while simultaneously offering some relief. And I sit on an airplane on my way to my second conference in two weeks, having received great news in the 4 days I was home, and yet I cry because I know that the relief I feel is temporary. Hopefully optimistic but not naive. Despite our best efforts this disease will progress. I can sit by the pool at my favorite hotel in Palo Alto so incredibly grateful for these people I've met because of this disease, and cry as I walk back to my room moments later cursing the world and wishing it all away in a trade for normalcy. I don't know normalcy though, I suppose. This is our normal. It's breathtaking to see the world in this way, a gift not many people get. And I love it and hate it equal parts.

Drew's lung function tanked this summer. We saw his doctors many times, and I used my personal network to crowd source ideas for what might be wrong. Early on, there were suggestions by my peers that a fungal infection might be the culprit. With no symptoms except for a rapidly declining FEV1, we spent countless hours troubleshooting - trying everything in our "evidence-based" bag of tricks. There's not significant evidence to suggest that fungus causes lung function decline in people with CF. Or I guess I should say that there's not peer-reviewed published scientific evidence. The "evidence" that I have, the experiences of the people living with cystic fibrosis who have lived through our same uncertain hell, is often discounted. And I get it, really. I am pro evidence-based medicine. I just can't help but think about the value of the individual human experience we are missing.

Despite there being no physical symptoms of exacerbation - data that I have based on our prior experiences, things that are in fact documented in his EHR - we opted for treating this like an exacerbation because we didn't know what else to do. The "we" I speak of is us and his care team. The first step was an oral antibiotic. Drew has never responded to oral antibiotics. I was skeptical of this plan but ultimately on board. After 10 days, there was actually increased decline. He has found benefit from steroids in the past due to the wonky nature of his airways, but I can identify with accuracy the situations that will be improved with steroids and this wasn't one of them. We tried anyway. The downside was outrageous mood swings. While medically induced, I still expect certain behaviors from my kids, and its absolutely heartbreaking to watch your 6yo, who feels like he's going to jump out of his skin and not understand why, sob in his bedroom for hours because, medication or not, hitting other people is not ok. The steroid may have slowed the decline, but his lung function continued to drop.

When we have to make the hard decisions about whether or not to try IV antibiotics, more frustration and disappointment came out of fear, from all of us, that yet another treatment, another procedure would render the same results and we would be left with even fewer options. It's so hard to balance these decisions. His health is a priority, but so is his life, and living means so much more than being alive. Can I take away the joy of the pool or the promised vacation with the hope that this could put us back on track? How could I convince him that this is the right thing to do when all of the preventative medicine he takes, the stuff we tell him we must do for hours every day to keep him healthy has left him so very sick. What the fuck kind of sick logic does this disease demand?

We opt for a combination of IV medications that we hope would target the bacteria that we suspect might be causing the problem, weighing dosing schedules against our ability to administer them from home because there's no way we are admitting him to the hospital when he looks and feels fine. We trade sleep for this decision and our whole world pays for it. Work commitments change and home routines suffer as patience wears thin. Tired and terrified while pretending to be neither is a bad combination. And yet we continue, a constant barrage of choices in our faces, very serious ones with real consequences.

They told us when he had his PICC placed that we might want to have a conversation with our team about a port given the increasing difficulty with placement of a central line due to the fact that he's had 9 in his 6 years of life and scar tissue is getting in the way. With some ativan to take the edge off, we both cried as they stuck him, then stuck him again. His fear has scars showing the distance he's come. I have fear too, but all of my scars are on the inside. I'm a little tougher to break, and each time we do this feels a little more routine, a little more scary.

The IV meds didn't help either. With our lung function creeping closer and closer to 70%, we re-group for another team meeting. I want to know all of the options. We decide on a bronchoscopy, something that will allow them to get a reasonable sample of what might be in his lungs and enable us to target our therapy. Why didn't we start here?  Everything comes with its own risks, and the care team was hopeful that we could address the problem more easily. I trust them and agreed with their plan. I also asked about the risks of treating with an anti-fungal medication. After all, we just followed the script for oral and IV antibiotics and also a steroid, each with its own risks, especially given that they seemed to be the wrong choice as they showed no clinical benefit and the risks of antibiotic resistance in this population is enormous. There is not evidence to support a fungal infection causing a decline in lung function, or "published, peer-reviewed, scientific evidence" anyway. There is a surplus of anecdotal experiential evidence.

With a family vacation just a day away, we go to the hospital for this outpatient procedure. I do my best distraction song and dance while they give him sleepy air. My husband and I don't even go to these together anymore because we've been through them so many times, and life and work don't pause for this disease. I anxiously wait to be called back to see the doctor in what I can only describe to be the purgatory that is the same day surgery waiting room. Her words have so much power to change my life and I'm terrified. But all looks good! His lungs look as good on the inside as they sounds on the outside and we are left wondering. We have to wait 5 long days to see what grows on the culture, to see what explanation we might get from another piece of data. I've convinced that I lose about a day of my life in this unavoidable worry that comes from this unavoidable waiting.

By now, his doctor has given me a prescription for the anti-fungal medication I've been asking for. I promised not to use it until after the bronch, We agree that we have nothing to lose by starting it while we go on our vacation and wait for our results. When the results do come back and show that he has a fungal infection, we exchange emails with his care team about their lack of optimism that the anti-fungal will have an impact given the type of fungus he has, a more common household name I'm told, but without other options it seems to be the last resort and we go forward.

After 4 weeks on the treatment, we wake up early, get our treatments done before the sun comes up, and head into the office for an assessment. I have a preference for the 7:30am  Monday morning appointment as I believe it lowers the cross-infection risk, something that no evidence exists for. It gives me some resolve to have this time slot and I'm grateful to the team that recognizes that and works to accommodate me. He's up to 84! It seems to be working! He's re-gained more 10pts of that lost lung functions and we are elated! We decide to stay the course and after another 4 weeks he is up to 91. I express my joy to his care team who is equally delighted, admitting their early skepticism about this being the cause of the problem and acknowledging the treatment as the reason for our success. I certainly didn't set out to prove anyone wrong, only to make him well, but it's nice to hear that they are learning along with me.

This is a happy story, but it's not the end of the story. We will have treatment decisions to make again next month - stay the course? Make a change? What changes will happen that will be out of our control? Will the next culture show a bug that's completely resistant? Is there anything I can do to prevent that? To protect him?

The past two courses of IV antibiotics that Drew has had have not been needed. We treated a bad case of acid reflux and a fungal infection with hard core antibiotics. We followed the guidelines and used decision trees and made the decisions that we felt we had to make, and we were wrong. Drew has been on inhaled antibiotics - 2 different medications alternated in 15 day cycles, inhaled 3x a day, everyday for the past 3 years. In those 3 years he has not had a bacterial infection in his lungs, an "exacerbation". In 3 years he has not gotten sick. He's been on IV and oral antibiotics that he hasn't needed, and been sedated countless times. He has been admitted for central line infections for central lines he hasn't needed, all because we followed the rules. I don't think we made a mistake, we did what we knew how to do. But now that we know better we need to do better. I wholeheartedly believe there has to be a better way. We might talk about the risks of exploring medications or procedures that lack "evidence", but I also want to talk about the risks of evidence based medications and treatments that we use when we don't need them. How might we get the right thing to the right person at the right time, every time? How might we improve the mental health of our patients and caregivers who are rightfully distraught over the lack of answers to what should be straightforward questions? Why is there still so much uncertainty in medicine? What responsibility should we put on people to advocate for themselves, tracking their own outcomes and then sharing them with the rest of the healthcare system to enable personalization of treatments, and then machine learning to aggregate all of these N of 1's, improving population health through the spread of personalized solutions.

Some new opportunities have recently come my way to improve things within the healthcare system. My goals remain the same - right person, right solution, right time, every time. I think success is more than just improved outcomes just as living is more than just being alive. Success for me is influence. When more people believe that this is possible, when they challenge the status quo and try out of the box solutions, that feels like success. I know I'm making a difference. We are farther in this culture change than we were 5 years ago when I realized my purpose in all of this. As I fly to another healthcare conference, my second in two weeks - a conference where patients and caregivers are being introduced to the world as a symbol of action and influence, a conference where our participation is being fully financially supported - I know that my role is to honestly and vulnerably share these stories and ideas for change. I've figured out how to fold the worst narrative of my life into triumph, and that for me is how I measure success.




          Comment on Lecture 1 | Machine Learning (Stanford) by Gurubux Gill      Cache   Translate Page      
Syllabus <br /><a href="http://cs229.stanford.edu/syllabus.html">http://cs229.stanford.edu/syllabus.html</a>
          Machine Learning/AI Engineer - Groom & Associes - Montréal, QC      Cache   Translate Page      
Experience with tensorflow or other backends, keras or other frameworks, scikit-learn, OpenCV, Pandas. An international company is looking for Machine Learning...
From Groom & Associes - Fri, 08 Mar 2019 21:29:13 GMT - View all Montréal, QC jobs
          Senior Data Scientist / Machine Learning Engineer - PubMatic - Montréal, QC      Cache   Translate Page      
Job Description We are looking for a strong Data Scientist or Machine Learning Engineer - a proven 'doer' to develop, implement and extend data-intensive...
From PubMatic - Tue, 04 Dec 2018 04:08:14 GMT - View all Montréal, QC jobs
          Data Scientist - Oliver Wyman - New York, NY      Cache   Translate Page      
Demonstrate solid and battle-tested understanding of the standard canon of machine learning practices, including but not limited to:....
From Marsh & McLennan Companies - Sat, 05 Jan 2019 15:04:15 GMT - View all New York, NY jobs
          Senior Data Scientist - Mischief - San Francisco, CA      Cache   Translate Page      
Familiarity with experimental design and A/B testing. We do this through intelligent design and machine learning applications that looks to tie together an...
From Mischief - Mon, 21 Jan 2019 19:50:03 GMT - View all San Francisco, CA jobs
          By: Andreas Karnthaler      Cache   Translate Page      
In my opinion pattern recognition will become even more important with the increasing use of Big Data and machine learning algorithms. Being able to go through a huge amount of statistical data and visualize them to find patterns will be a powerful tool for the "explain" part as described in Question 2.
          Machine Learning - Al Manal Training Center , United Arab Emirates, Abu Dhabi,Abu Dhabi       Cache   Translate Page      
Machine Learning Training:

Machine learning! It’s a branch of computer science. It’s one of the most learnt courses today as the opportunities 
are abundant. If you want to learn Machine learning course in Abu Dhabi, then reach Al Manal Training ;
As new technologies evolved, machine learning was altered greatly and our Machine learning classes in Abu Dhabi 
will give you a clear understanding.

Below are some of the topics that we will discuss in our machine learning training classes at our institute:

Machine Learning
Data Preprocessing
Introduction to Supervised Learning
Simple and Multiple Linear Regression
Polynomial Regrssion

Linear Methods for Classification
Logistic Regression
K-Nearest Neighbours
Support Vector Machines
Kernel SVM
Naive Bayes
Decision Tree
Random Forest

Introduction to Unsupervised Learning
Cluster Analysis
K Means Clustering

Reinforcement Learning

Natural Language Processing

Deep Learning
Artificial Neural Networks

Dimensionality Reduction

Model Selection Procedures

Cost: 3500 AED

Duration: Upto 30 Hours


          Data Science - Al Manal Training Center , United Arab Emirates, Abu Dhabi,Abu Dhabi       Cache   Translate Page      
Data Science Training:

Data Science deals with leveraging the data, which is extracted for making some good decisions in a ;
Usually data is in structured and unstructured format and data analysts leverage their skills in maths, programming, 
and statistics to organize and clean the ;
So, to be proficient in it, get Data Science course in Abu Dhabi at Al Manal training institute.


Course Outline:


Data Science:
Statistics
Probability
Linear Algebra
Calculus
Python Programming
Introduction to Data Science
Data Cleaning using python
Data Visualization using python
Data Modelling using python
Machine Learning using python

Cost: 3800 AED

Duration: Upto 40 Hours


          Data Analytics - Al Manal Training Center , United Arab Emirates, Abu Dhabi,Abu Dhabi       Cache   Translate Page      
Data Analytics Training:

Data Analytics deals with leveraging the data, which is extracted for making some good decisions in a ;
Usually data is in structured and unstructured format and data analysts leverage their skills in maths, programming, 
and statistics to organize and clean the ;
So, to be proficient in it, get Data analytics course in Abu Dhabi at Al Manal training institute.


Course Outline:


Python:

Introduction to Python
Datatypes
Lists, Ranges & Tuples
Dictionaries & Sets
Control flow statements
Loops in Python
Functions & Modules 
Object Oriented Concepts 
Using databases in Python

Data Analytics:

Introduction to Data Analytics
Data Cleaning using python
Data Visualization using python
Data Modelling using python
Machine Learning using python

Cost: 3000 AED

Duration: Upto 40 Hours


          Data Engineer (Data Warehouse) - Bandwidth - Raleigh, NC      Cache   Translate Page      
Machine Learning &amp; statistics experience. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Mon, 25 Feb 2019 20:43:17 GMT - View all Raleigh, NC jobs
          Exacerbations in Chronic Obstructive Pulmonary Disease: Identification and Prediction Using a Digital Health System      Cache   Translate Page      
Background: Chronic obstructive pulmonary disease (COPD) is a progressive, chronic respiratory disease with a significant socioeconomic burden. Exacerbations, the sudden and sustained worsening of symptoms, can lead to hospitalization and reduce quality of life. Major limitations of previous telemonitoring interventions for COPD include low compliance, lack of consensus on what constitutes an exacerbation, limited numbers of patients, and short monitoring periods. We developed a telemonitoring system based on a digital health platform that was used to collect data from the 1-year EDGE (Self Management and Support Programme) COPD clinical trial aiming at daily monitoring in a heterogeneous group of patients with moderate to severe COPD. Objective: The objectives of the study were as follows: first, to develop a systematic and reproducible approach to exacerbation identification and to track the progression of patient condition during remote monitoring; and second, to develop a robust algorithm able to predict COPD exacerbation, based on vital signs acquired from a pulse oximeter. Methods: We used data from 110 patients, with a combined monitoring period of more than 35,000 days. We propose a finite-state machine–based approach for modeling COPD exacerbation to gain a deeper insight into COPD patient condition during home monitoring to take account of the time course of symptoms. A robust algorithm based on short-period trend analysis and logistic regression using vital signs derived from a pulse oximeter is also developed to predict exacerbations. Results: On the basis of 27,260 sessions recorded during the clinical trial (average usage of 5.3 times per week for 12 months), there were 361 exacerbation events. There was considerable variation in the length of exacerbation events, with a mean length of 8.8 days. The mean value of oxygen saturation was lower, and both the pulse rate and respiratory rate were higher before an impending exacerbation episode, compared with stable periods. On the basis of the classifier developed in this work, prediction of COPD exacerbation episodes with 60%-80% sensitivity will result in 68%-36% specificity. Conclusions: All 3 vital signs acquired from a pulse oximeter (pulse rate, oxygen saturation, and respiratory rate) are predictive of COPD exacerbation events, with oxygen saturation being the most predictive, followed by respiratory rate and pulse rate. Combination of these vital signs with a robust algorithm based on machine learning leads to further improvement in positive predictive accuracy. Trial Registration: International Standard Randomized Controlled Trial Number (ISRCTN): 40367841; http://www.isrctn.com/ISRCTN40367841 (Archived by WebCite at http://www.webcitation.org/6olpMWNpc)

Views: 0
          Detecting Hypoglycemia Incidents Reported in Patients’ Secure Messages: Using Cost-Sensitive Learning and Oversampling to Reduce Data Imbalance      Cache   Translate Page      
Background: Improper dosing of medications such as insulin can cause hypoglycemic episodes, which may lead to severe morbidity or even death. Although secure messaging was designed for exchanging nonurgent messages, patients sometimes report hypoglycemia events through secure messaging. Detecting these patient-reported adverse events may help alert clinical teams and enable early corrective actions to improve patient safety. Objective: We aimed to develop a natural language processing system, called HypoDetect (Hypoglycemia Detector), to automatically identify hypoglycemia incidents reported in patients’ secure messages. Methods: An expert in public health annotated 3000 secure message threads between patients with diabetes and US Department of Veterans Affairs clinical teams as containing patient-reported hypoglycemia incidents or not. A physician independently annotated 100 threads randomly selected from this dataset to determine interannotator agreement. We used this dataset to develop and evaluate HypoDetect. HypoDetect incorporates 3 machine learning algorithms widely used for text classification: linear support vector machines, random forest, and logistic regression. We explored different learning features, including new knowledge-driven features. Because only 114 (3.80%) messages were annotated as positive, we investigated cost-sensitive learning and oversampling methods to mitigate the challenge of imbalanced data. Results: The interannotator agreement was Cohen kappa=.976. Using cross-validation, logistic regression with cost-sensitive learning achieved the best performance (area under the receiver operating characteristic curve=0.954, sensitivity=0.693, specificity 0.974, F1 score=0.590). Cost-sensitive learning and the ensembled synthetic minority oversampling technique improved the sensitivity of the baseline systems substantially (by 0.123 to 0.728 absolute gains). Our results show that a variety of features contributed to the best performance of HypoDetect. Conclusions: Despite the challenge of data imbalance, HypoDetect achieved promising results for the task of detecting hypoglycemia incidents from secure messages. The system has a great potential to facilitate early detection and treatment of hypoglycemia.

Views: 0
          Azure Databricks bekommt neue Funktionen      Cache   Translate Page      
Microsoft spendiert Azure Databricks neue Funktionen: Data Engineering Light, Vorschau für MLflow und Integration mit Azure Machine Learning. Mit Azure Databricks können Kunden in wenigen Minuten eine Apache Spark-Umgebung einrichten. Die native Integration Mehr…...(read more)
          Azure Monitor: Schwellenwerte für Metriken mit Machine Learning festlegen       Cache   Translate Page      
Für Azure Monitor stehen ab sofort Metrikwarnungen mit dynamischen Schwellenwerten in einer Public Preview zur Verfügung. Dank der dynamischen Schwellenwerte müssen Nutzer die Schwellenwerte für Warnhinweise nicht mehr manuell identifizieren und festlegen Mehr…...(read more)
          Contact center: i cinque fattori che migliorano produttività e customer experience      Cache   Translate Page      

Cisco ha condiviso i cinque fattori che trasformeranno i contact center e la customer experience nei prossimi cinque anni.

Secondo l’azienda le aziende che mettono i loro clienti al centro di tutto ciò che fanno possono trasformare il loro business. Nonché la la customer experience dei loro clienti. Tale trasformazione non avviene però da un giorno all’altro. Richiede invece una visione chiara, cambiamenti culturali e uno sguardo approfondito alla tecnologia utilizzata dall’azienda e dai clienti.

Tutte le relazioni solide, sottolinea Cisco, si basano su una comunicazione sana e sulla comprensione reciproca. Questo non è diverso per quel che riguarda i clienti. Per costruire relazioni più profonde e significative, l’azienda deve davvero ascoltare e comprendere i clienti.

Nell’attuale scenario popolato da così tante tecnologie, piattaforme e canali differenti, le aspettative sono però molto cambiate. Per comunicare in modo efficace ora è necessario trasformare le attività con tecnologie e processi moderni. Progettati per dialogare con i clienti ovunque si trovino.

Percorsi complicati che impediscono ai clienti di accedere alle risorse e alle informazioni di cui hanno bisogno, non funzionano. È vero che il contact center non è l’unico touchpoint con il cliente lungo il percorso. Però, esso svolge un ruolo enorme nel forgiare relazioni più profonde con il cliente e per l’azienda.

1. L’importanza della customer experience

Secondo Nicholas Webb, esperto di innovazione, attualmente ci troviamo nell’economia dell’esperienza. La customer experience attraverso tutti i punti di contatto, nei canali digitali e non, è ciò che determina il successo di un’azienda.

Cisco aggiunge che ci sono tre elementi principali che modellano la customer experience dei clienti di un’azienda. Innanzitutto i dati, da una moltitudine di sistemi e applicazioni, che forniscono informazioni fondamentali sulla customer experience.

Sull’esperienza dei clienti mentre navigano in un processo spesso multi-stop all’interno del brand. Comprendere e analizzare i dati può dire molto all’azienda su come i clienti stanno vivendo il marchio. In modo che possano essere intraprese azioni per migliorare il customer journey. Ciò porta alla brand loyalty e alla customer retention.

Gli agenti, sia digitali che umani, fanno la differenza nella creazione di interazioni significative con i clienti. Svolgono un ruolo chiave che è direttamente correlato a come i clienti prendono decisioni d’acquisto. Investire in un client moderno, intuitivo, omnicanale, e abilitato dall’intelligenza artificiale contribuirà notevolmente a migliorare la produttività e la job satisfaction dell’agente.

I canali che i clienti utilizzano per comunicare con l’azienda, una volta in silos, ostacolano la produttività degli agenti. Nonché la possibilità per i clienti di interagire facilmente con il marchio. Un’esperienza omnicanale attentamente integrata consente di ottimizzare la forza lavoro dell’azienda. Offrendo, al contempo, un’esperienza fluida con la possibilità di creare engagement altamente personalizzati per il cliente.

2. Il percorso verso il cloud

La transizione sul cloud dei contact center on-premise crea enormi opportunità di crescita e innovazione. Ma può anche creare interruzioni che la maggior parte delle aziende semplicemente non può permettersi. Scegliere un approccio graduale è il modo più razionale per fare questa importante transizione.

Il contact center è una delle applicazioni più interconnesse in azienda, con dozzine di integrazioni a più sistemi. Questo è uno dei motivi per cui la penetrazione del cloud non ha tenuto il passo di altre applicazioni quali email, CRM ed ERP.

I servizi cloud ibridi sono un modo intelligente per iniziare a beneficiare dell’innovazione del cloud continuando a eseguire le operazioni critiche senza interruzioni. Adottando e integrando moderni servizi cloud e tecnologie quali analytics e intelligenza artificiale alle implementazioni on-premise, si può iniziare a potenziare gradualmente con il cloud il proprio contact center. Allo scopo di acquisire i vantaggi in modo rapido, cost-effective e con il rischio minimo.

3. Intelligenza artificiale

L’utilizzo di tecnologie di intelligenza artificiale e machine learning consente di fornire servizi self-service e assistenti virtuali ai clienti. Ciò aiuta gli agenti del contact center con contesto, cognizione e intelligenza in tempo reale. Questo è uno dei modi più efficaci per rendere la vita lavorativa degli operatori più facile e risolvere il sovraccarico di informazioni.

L’intelligenza artificiale rimuove le attività banali e garantisce agli agenti informazioni sempre a portata di mano. In modo tale che essi possano prendersi cura di ciascun cliente a un livello altamente individuale e personalizzato.

4. Cloud analytics

Sono molti i dati sui clienti, ma il problema è che sono da diverse fonti, in diversi formati e gestiti da singole business unit. Ciò rende difficile ottenere una visione unica del cliente.

La soluzione è consolidare i dati da tutte queste fonti ed estrarre il significato da questi dati. L’unico modo per analizzare questa abbondanza di dati inestimabili è mediante il reporting di cloud analytics. Ciò fornirà preziose informazioni commerciali e una visione completa del customer journey, in tempo reale e nello storico. In questa maniera l’azienda potrà migliorare l’efficienza operativa, le prestazioni finanziarie e le interazioni con i clienti in modi innovativi.

5. Rimuovere i silos

Gli agenti del contact center necessitano di essere strettamente connessi con il resto dell’organizzazione. Il modo migliore per farlo è di fornire loro un accesso rapido e facile agli esperti, mediante sistemi di unified communications (UC) e team collaboration.

È necessario aiutare gli agenti a raggiungere chiunque nell’organizzazione tramite chat o email o chiamandoli ovunque si trovino. In questo modo l’azienda non solo rende gli agenti più efficienti, migliora anche la customer experience dei clienti, e loro lo noteranno.

Maggiori informazioni sono disponibili sul sito dell’azienda, a questo link.

L'articolo Contact center: i cinque fattori che migliorano produttività e customer experience è un contenuto originale di 01net.


          Continuous Testing per aziende che rilasciano una build al giorno      Cache   Translate Page      

Una ricerca condotta da Capgemini e Sogeti in collaborazione con Broadcom, fa emergere che il Continuous Testing, ossia ovvero il processo che prevede una validazione veloce ed efficiente delle nuove versioni dei software in ambienti agile per il tramite di test altamente automatizzati, si sta affermando nelle grandi aziende.

Il 32% degli IT manager afferma che le loro divisioni hanno “completamente abbracciato il Continuous Testing e il 58% del campione implementa una nuova build su base giornaliera (il 26% una ogni ora).

In cerca di maggiore automazione

Metodologia dell’indagine

Il Continuous Testing Report (CTR) 2019 riunisce i dati delle indagini e i contributi di esperti del settore al fine di delineare le potenziali sfide e gli approcci per trasformare le pratiche di test nell’era di agile e DevOps. Si basa sulle opinioni di diversi esperti del settore di Capgemini, Sogeti e Broadcom, supportate dai risultati di un’indagine globale condotta con 500 interviste a IT manager con elevata seniority di aziende medio-grandi (con oltre 1.000 dipendenti) operanti in diversi comparti, tra cui Financial Services, High Tech, Healthcare and Life Sciences, Telecommunications, Media and Entertainment e Manufacturing in otto paesi diversi.

C’è però ancora spazio di miglioramento per ottimizzare i processi di Continuous Testing: l’automazione viene utilizzata per eseguire solo il 24% dei test, il 24% degli scenari di business end-to-end e per generare il 25% dei dati richiesti per i test.

Un uso maggiore dell’automazione potrebbe migliorare la velocità delle attività di testing dei team agile: il 36% degli intervistati ha dichiarato che più del 50% dei tempi di collaudo è dedicato a ricercare, gestire, mantenere e generare dati per i test.

La discrezionalità con cui i team di lavoro si sono mossi in autonomia, ha avuto come conseguenza che molte imprese si siano trovate a dovere gestire situazioni fuori controllo, con un’ampia varietà di approcci alla Quality Assurance e al test automation.

Emerge la shadow quality

Secondo lo studio per riprendere il controllo le imprese devono indirizzarsi verso strumenti e metodologie che garantiscano un controllo centralizzato della qualità da parte dei team agile, attraverso linee guida più chiare in ambito Quality Assurance ed il supporto di tecnologie Quality Assurance smart.

Uno sviluppo che può portare a risultati promettenti è rendere l’orchestrazione e l’esecuzione dei test molto più smart attraverso l’utilizzo di tecnologie di intelligenza artificiale.

Con l’introduzione di funzionalità di machine learning, i sistemi saranno in grado di determinare automaticamente i test necessari nei cicli di rilascio e di produzione.

Il report ha evidenziato la necessità di migliorare trasparenza e gestione dei test agile. Per il 35% dei dirigenti intervistati, gli elementi fondanti per validare la capacità di gestione dell’orchestrazione dei test e dei rilasci sono “una tracciatura completa delle attività di test svolte” e un “processo consolidato di test e rilascio”. Il 32% del campione, invece, ha sottolineato la necessità di uno “spazio condiviso per la collaborazione tra i team” e di una “visibilità continua dei test e rilasci in esecuzione”.

Se ci si sofferma sulle sfide degli ambienti di test, la carenza di una struttura centralizzata finalizzata alla fornitura di un ambiente ready-to-use è evidente.

I team sprecano troppo tempo per ottenere ambienti di collaudo completi, tanto che quattro intervistati su dieci (40%) hanno dichiarato che i loro team dedicano più della metà del loro tempo alla creazione e manutenzione degli ambienti di test.

Sviluppo e test: team interdisciplinari

Questo spinge Francesco Fantazzini, Technology & Innovation Lead, Capgemini Business Unit Italy a dire che “I prossimi due o tre anni saranno cruciali per il Continuous Testing – dato che le aziende dovranno – risolvere il dilemma della transizione verso team sempre più autonomi, in cui tutti si devono sentire responsabili della qualità“.

Negli ultimi anni i ruoli degli sviluppatori e degli addetti ai test si sono evoluti in modo significativo. Secondo lo studio attualmente gli sviluppatori sono molto più vicini al cliente, con un ruolo di primo piano nel definire la user experience, mentre gli addetti ai test hanno abbandonato una logica di lavoro a silos, collaborando con sviluppatori e team aziendali, e ciò significa che vengono coinvolti molto prima nel processo di sviluppo.

I ruoli e le responsabilità di sviluppatori e addetti ai test si stanno parzialmente mischiando, tuttavia è fondamentale avere esperti focalizzati sulla Quality Assurance e sui test all’interno dei team agile.

Il report prosegue sottolineando che, se da un lato la creazione di team interdisciplinari rappresenta un passo avanti, dall’altro crea delle sfide. È necessario infatti che ogni membro del team abbia una comprensione olistica dell’intero processo, con gli addetti ai test che devono aggiornare le proprie competenze tecniche.

Le aziende devono soddisfare i requisiti di aggiornamento delle competenze e utilizzare un nuovo approccio integrato per raggiungere veramente il pieno potenziale del Continuous Testing.

Continuous testing

L'articolo Continuous Testing per aziende che rilasciano una build al giorno è un contenuto originale di 01net.


          Alphabet Launches Chrome Extension That Filters Comments With AI      Cache   Translate Page      
Following up on the "Perspective" hate speech filtering experiment from 2017, one Alphabet's subsidiaries, Jigsaw, recently released a machine learning-powered Chrome extension designed to filter out "toxic" comments on high traffic sites. Out of curiosity, I downloaded the extension on a fresh Chrome install, and found that it features a virtual nob that lets users tune the "volume" of the comments sections in YouTube, Facebook, Twitter, Reddit, and Disqus comment sections. Twisting the knob gradually filters out more and more comments in real time. As the developers note, it definitely misses some nasty comments while hiding other comments that aren't particularly "toxic" at all, but based on my quick test with some controversial YouTube videos, the sheer variety of language it can seemingly interpret is remarkable. The machine learning powering Tune is experimental. It still misses some toxic comments and incorrectly hides some non-toxic comments. We're constantly working to improve the underlying technology, and users can easily give feedback right in the tool to help us improve our algorithms. Tune isn't meant to be a solution for direct targets of harassment (for whom seeing direct threats can be vital for their safety), nor is Tune a solution for all toxicity. Rather, it's an experiment to show people how machine learning technology can create new ways to empower people as they read discussions online. Discussion
          Senior Machine Learning Engineer - Fluke - Everett, WA      Cache   Translate Page      
Design, implement, deploy and analyze machine learning systems for machine health, including algorithms, datasets, training and performance....
From Fluke - Mon, 17 Dec 2018 22:18:16 GMT - View all Everett, WA jobs
          AI Platform Software Architect - DISCO - Austin, TX      Cache   Translate Page      
Should have prior experience designing, implementing, and operating scalable distributed systems for Big Data, Machine Learning or Machine Learning Problems....
From Disco - Tue, 12 Mar 2019 16:09:59 GMT - View all Austin, TX jobs
          Information Design and Visualization Specialist - Booz Allen Hamilton - Washington, DC      Cache   Translate Page      
Are you fascinated by the possibilities presented by the IoT, machine learning, and artificial intelligence advances?...
From Booz Allen Hamilton - Thu, 08 Nov 2018 16:40:47 GMT - View all Washington, DC jobs
          Solutions Architect, Accelerated Computing - NVIDIA - Santa Clara, CA      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 - Thu, 22 Nov 2018 07:57:08 GMT - View all Santa Clara, CA jobs
          Director of Sales, Global AI/DL Data Center Strategy - NVIDIA - Santa Clara, CA      Cache   Translate Page      
Demonstrated the work ethic to lead at NVIDIA. Machine learning, data analytics, and artificial intelligence experience preferred....
From NVIDIA - Mon, 01 Oct 2018 20:01:50 GMT - View all Santa Clara, CA jobs
          Machine Learning and Big Data Engineer - Makrwatch - Cali, Valle del Cauca      Cache   Translate Page      
We believe video will grow exponentially over the coming years and content creators will continue to challenge mass media as we know it.... $4.000.000 - $9.000.000 al mes
De Indeed - Fri, 15 Feb 2019 18:38:03 GMT - Ver todos: empleos en Cali, Valle del Cauca
          Siber güvenlikte makine öğrenimi dönemine girdik      Cache   Translate Page      

Önümüzdeki dönemde siber saldırılarda otomasyon daha üst seviyeye çıkacak. Siber suçlular, veri toplama girişimlerinde yapay zeka (AI-Artificial Intelligence) sistemlerini ve makine öğrenimi (ML-Machine Learning) uygulamalarını artık daha fazla kullanacak. Böylece daha kişiselleştirilmiş ve sofistike kimlik avı saldırıları yapabilecekler. Önümüzdeki dönemde siber saldırılarda otomasyon daha üst seviyeye çıkacak. Siber suçlular, veri toplama girişimlerinde yapay zeka (AI-Artificial […]

The post Siber güvenlikte makine öğrenimi dönemine girdik appeared first on ABC Gazetesi.


          Data Scientist Lead - Schneider National - Green Bay, WI      Cache   Translate Page      
Experience with machine learning software (e.g., R, Python, SPSS, SAS), data access/manipulation (e.g., SQL, pandas, dplyr) and NoSQL databases (e.g., MongoDB,...
From Schneider National - Thu, 03 Jan 2019 06:22:26 GMT - View all Green Bay, WI jobs
          Data Scientist Lead - Schneider - Green Bay, WI      Cache   Translate Page      
Experience with machine learning software (e.g., R, Python, SPSS, SAS), data access/manipulation (e.g., SQL, pandas, dplyr) and NoSQL databases (e.g., MongoDB,...
From Schneider - Wed, 02 Jan 2019 23:36:22 GMT - View all Green Bay, WI 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 - Tue, 25 Dec 2018 09:45:46 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 - Tue, 25 Dec 2018 09:45:46 GMT - View all Palo Alto, CA jobs
          Safe Convex Learning under Uncertain Constraints. (arXiv:1903.04626v1 [math.OC])      Cache   Translate Page      

Authors: Ilnura Usmanova, Andreas Krause, Maryam Kamgarpour

We address the problem of minimizing a convex smooth function $f(x)$ over a compact polyhedral set $D$ given a stochastic zeroth-order constraint feedback model. This problem arises in safety-critical machine learning applications, such as personalized medicine and robotics. In such cases, one needs to ensure constraints are satisfied while exploring the decision space to find optimum of the loss function. We propose a new variant of the Frank-Wolfe algorithm, which applies to the case of uncertain linear constraints. Using robust optimization, we provide the convergence rate of the algorithm while guaranteeing feasibility of all iterates, with high probability.


          Accelerated Learning in the Presence of Time Varying Features with Applications to Machine Learning and Adaptive Control. (arXiv:1903.04666v1 [math.OC])      Cache   Translate Page      

Authors: Joseph E. Gaudio, Travis E. Gibson, Anuradha M. Annaswamy, Michael A. Bolender

Features in machine learning problems are often time varying and may be related to outputs in an algebraic or dynamical manner. The dynamic nature of these machine learning problems renders current accelerated gradient descent methods unstable or weakens their convergence guarantees. This paper proposes algorithms for the case when time varying features are present, and demonstrates provable performance guarantees. We develop a variational perspective within a continuous time algorithm. This variational perspective includes, among other things, higher-order learning concepts and normalization, both of which stem from adaptive control, and allows stability to be established for dynamical machine learning problems. These higher-order algorithms are also examined for achieving accelerated learning in adaptive control. Simulations are provided to verify the theoretical results.


          Senior AI Solutions Architect - Industrial - Petuum - Sunnyvale, CA      Cache   Translate Page      
Machine learning or IIoT preferred. PaaS, SaaS, IaaS and business intelligence/analytics implementation experience are a plus....
From Petuum - Sun, 06 Jan 2019 08:08:10 GMT - View all Sunnyvale, CA jobs
          An Efficient Augmented Lagrangian Based Method for Constrained Lasso. (arXiv:1903.05006v1 [math.OC])      Cache   Translate Page      

Authors: Zengde Deng, Anthony Man-Cho So

Variable selection is one of the most important tasks in statistics and machine learning. To incorporate more prior information about the regression coefficients, the constrained Lasso model has been proposed in the literature. In this paper, we present an inexact augmented Lagrangian method to solve the Lasso problem with linear equality constraints. By fully exploiting second-order sparsity of the problem, we are able to greatly reduce the computational cost and obtain highly efficient implementations. Furthermore, numerical results on both synthetic data and real data show that our algorithm is superior to existing first-order methods in terms of both running time and solution accuracy.


          Selecting Optimal Parameters for XGBoost Model Training      Cache   Translate Page      
There is always a bit of luck involved when selecting parameters for Machine Learning model training. Lately, I work with gradient boosted trees and XGBoost in particular. We are using XGBoost in the enterprise to automate repetitive human tasks. While training ML models with XGBoost, I created a pattern to choose parameters, which helps me […]
          Google moves into the hotel booking space      Cache   Translate Page      
Google blog: “…Set on your flight but need to whittle down your hotel options? Let’s imagine you’re going to Miami at the end of March, and there are over 300 hotel results for your search. To help you find the right hotel for your trip, apply our new “Deals” filter. This filter uses machine learning to highlight hotels where one or more of our partners offer rates that are significantly lower than the usual price for that hotel or similar hotels nearby.  You can also view a hotel’s highlights—like a fancy pool, if it’s a luxury hotel, or if it’s popular with families—with expanded pages for photos and reviews curated with machine learning… Google Hotels
          Lead Software Engineer - Java - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Streaming....
From Capital One - Thu, 28 Feb 2019 18:12:30 GMT - View all Tysons Corner, VA jobs
          Principal Associate Software Engineer - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), API Gateway, AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Kafka,...
From Capital One - Sat, 02 Feb 2019 15:41:05 GMT - View all Tysons Corner, VA jobs
          AI Platform Software Architect - DISCO - Austin, TX      Cache   Translate Page      
Should have prior experience designing, implementing, and operating scalable distributed systems for Big Data, Machine Learning or Machine Learning Problems....
From Disco - Tue, 12 Mar 2019 16:09:59 GMT - View all Austin, TX 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 - Thu, 20 Dec 2018 07:48:32 GMT - View all Providence, RI jobs
          I see artificial people      Cache   Translate Page      
When people think of artificial intelligence, AI, they think of Alexa, Siri, Google Home and self-driving cars. When an AI dreams of humans it dreams up their faces. No really. Generative Adversarial Networks (GAN) are a relatively new concept in Machine Learning, introduced for the first time in 2014. Their goal is to synthesize artificial samples, such […]
           It takes more brain power to forget something than to remember it       Cache   Translate Page      

The team used fMRI and machine learning to track brain function#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

An important part of the human brain has to work harder to actively forget a memory than it does to remember it, according to the results of a newly-published study. The research is a step towards understanding how and why the brain is able to discard an experience, and could one day lead to a treatment designed to remove painful memories.

.. Continue Reading It takes more brain power to forget something than to remember it

Category: Science

Tags:
          Machine Learning / Atificial Intelligence Engineer - Eigen - Grass Valley, CA      Cache   Translate Page      
Worldwide, Eigen has deployed two medical devices Artemis a 3D semi-robotic prostate biopsy system and ProFuse an MRI image fusion product....
From Indeed - Mon, 10 Dec 2018 18:58:13 GMT - View all Grass Valley, CA jobs
          Senior AI/Deep Learning Software Engineer - St Josephs Hospital and Medical Center - Phoenix, AZ      Cache   Translate Page      
Ability to align business needs to development and machine learning or artificial intelligence solutions. Experience in natural language understanding, computer...
From Dignity Health - Tue, 27 Nov 2018 03:06:49 GMT - View all Phoenix, AZ jobs
          Beware the data science pin factory: The power of the full-stack data science generalist and the perils of division of labor through function      Cache   Translate Page      

assembly line
[This is a slightly modified version of the article I wrote for HBR originally published here].

In The Wealth of Nations, Adam Smith demonstrates how the division of labor is the chief source of productivity gains using the vivid example of a pin factory assembly line: “One [person]1 draws out the wire, another straights it, a third cuts it, a fourth points it, a fifth grinds it.” With specialization oriented around function, each worker becomes highly skilled in a narrow task leading to process efficiencies. Output per worker increases many fold; the factory becomes extremely efficient at producing pins.

This division of labor by function is so ingrained in us even today that we are quick to organize our teams accordingly. Data science is no exception. An end-to-end algorithmic business capability requires many data functions and companies usually create teams of specialists: research scientist, data engineers, machine learning engineers, causal inference scientists, and so on. Specialists’ work is coordinated by a product manager, with hand-offs between the functions in a manner resembling the pin factory: “one person sources the data, another models it, a third implements it, a fourth measures it” and on and on.

Alas, we should not be optimizing our data science teams for productivity gains; that is what you do when you know what it is you’re producing—pins or otherwise—and are merely seeking incremental efficiencies. The goal of assembly lines is execution. We know exactly what we want—pins in Smith’s example, but one can think of any product or service in which the requirements fully describe all aspects of the product and its behavior. The role of the workers is then to execute on those requirements as efficiently as possible.

But the goal of data science is not to execute. Rather, the goal is to learn and develop profound new business capabilities. Algorithmic products and services like recommendations systems, client engagement bandits, style preference classification, size matching, fashion design systems, logistics optimizers, seasonal trend detection, and more can’t be designed up-front. They need to be learned. There are no blueprints to follow; these are novel capabilities with inherent uncertainty. Coefficients, models, model types, hyper parameters, all the elements you’ll need must be learned through experimentation, trial and error, and iteration. With pins, the learning and design are done up-front, before you produce them. With data science, you learn as you go, not before you go.

In the pin factory, when learning comes first we do not expect, nor do we want, the workers to improvise on any aspect the product, except to produce it more efficiently. Organizing by function makes sense since task specialization leads to process efficiencies and production consistency (no variations in the end product).

But when the product is still evolving and the goal is to learn, specialization hinders our goals in several ways:

1. It increases coordination costs. Those are the costs that accrue in time spent communicating, discussing, justifying, and prioritizing the work to be done. These costs scale super-linearly with the number of people involved.2 When data scientists are organized by function the many specialists needed at each step, and with each change, and each handoff, and so forth, make coordination costs high. For example, a data science specialists focused on statistical modeling will have to coordinate with a data engineer any time a dataset needs to be augmented in order to experiment with new features. Similarly, any time new models are trained the research scientist will have to coordinate with a machine learning engineer to deploy them to production, etc. These coordination costs act as a tax on iteration and can hamper learning.

2. It exacerbates wait-time. Even more nefarious than coordinate costs is the time that elapses between work. While coordination costs can typically be measured in hours—the time it takes to hold meetings, discussions, design reviews—wait-times are commonly measured in days or weeks or even months! Schedules of functional specialists are difficult to align as each specialist is allocated to several initiatives. A one-hour meeting to discuss changes may take weeks to line up. And, once aligned on the changes, the actual work itself also needs to be scheduled in the context of multiple other projects vying for specialists’ time. Work like code changes or research that requires just a few hours or days to complete still may sit undone much longer before the resources are available. Until then, iteration and learning languish.

3. It narrows context. Division of labor can artificially limit learning by rewarding people for staying in their lane. For example, the research scientist who is relegated to stay within her function will focus her energy towards experimenting with different types algorithms: gradient boosting, neural nets, random forest, and so on. To be sure, good algorithm choices could lead to incremental improvements. But there is usually far more to gain from other activities like integrating new data sources. Similarly, she may develop a model that exhausts every bit of explanatory power inherent to the data. Yet, her biggest opportunity may lie in changing the objective function or relaxing certain constraints. This is hard to see or do when her job function is limited. Since the research scientist is specialized in optimizing algorithms, she’s far less likely to pursue anything else, even when it carries outsized benefits.

Telling symptoms can surface when data science teams are run like pin factories, for example in simple status updates: “waiting on ETL changes” and “waiting on ML Eng resources” are common blockers. However, I believe the more insidious impact lies in what you don’t hear, because you can’t lament what you haven’t yet learned. Perfect execution on requirements and complacency brought on by achieving process efficiencies can mask the difficult truth, that the organization is blissfully unaware of the valuable learnings they are missing out on.

The solution to this problem is, of course, to get rid of the pin factory. In order to encourage learning and iteration, data science roles need to be made more general, with broad responsibilities agnostic to technical function. That is, organize the data scientists such that they are optimized to learn. This means hiring “full stack data scientists”—generalists—that can perform diverse functions: from conception to modeling to implementation to measurement. With fewer3 people to keep in the loop, coordination costs plummet. The generalist moves fluidly between functions, extending the data pipeline to add more data, trying new features in the model, deploying new versions to production for causal measurement, and repeating the steps as quickly as new ideas come to her. Of course, the generalist performs the different functions sequentially rather than in parallel—she is just one person after all. However, doing the work typically takes just a fraction of the wait-time it would take for another specialist resource to come available. So, iteration time goes down.

Our generalist may not be as adept as a specialist in any one function. But we are not seeking functional excellence or small incremental improvements. Rather, we seek to learn and discover all-new business capabilities with step-change impact. With full context for the holistic solution she sees opportunities that a narrow specialist won’t. She has more ideas and tries more things. She fails more, too. However, the cost of failure is low and the benefits of learning are high. This asymmetry favors rapid iteration and rewards learning.

It is important to note that this amount of autonomy and diversity in skill granted to the full-stack data scientists depends greatly on the assumption of a solid data platform on which to work. A well constructed data platform abstracts the data scientists from the complexities of containerization, distributed processing, automatic failover, and other advanced computer science concepts. In addition to abstraction, a robust data platform can provide seamless hooks into an experimentation infrastructure, automate monitoring and alerting, provide auto-scaling, and enable visualization of debugging output and algorithmic results. These components are designed and built by data platform engineers, but to be clear, there is not a hand-off from the data scientist to a data platform team. It’s the data scientist that is responsible for all the code that is deployed to run on top of the platform. And, for the love of everything sacred and holy in the profession, don’t hand-off ETL for engineers to write.

I too was once lured to a function-based division of labor by the attraction of process efficiencies. But, through trial and error (is there no better way to learn?) I’ve found that more generalized roles better facilitate learning and innovating,4 and provide the right kinds of scaling: to discover and build many more business capabilities than a specialist approach. And, while there are some important considerations5 that may make this approach to organization more or less tenable in some companies (see footnote), I believe the full stack data scientist model provides a better starting place. Start with them, and then consciously (grudgingly) move toward a function-based division of labor only when clearly necessary.

Final thought.

There is further downside to functional specialization. It can lead to loss of accountability and passion from the workers. Smith himself criticizes the division of labor, suggesting that it leads to the dulling of talent—that workers become ignorant and insular as their roles are confined to a few repetitive tasks.6 While specialization may provide process efficiencies it is less likely to inspire workers.

By contrast, generalist roles provide all the things that drive job satisfaction: autonomy, mastery, and purpose.7 Autonomy in that they are not dependent on someone else for success. Mastery in that they know the business capability from end-to-end. And, purpose in that they have a direct connection to the impact on the business they’re making. If we succeed in getting people to be passionate about their work and making a big impact on the company, then the rest falls into place naturally.

Footnotes and References

[1]↩ I took the liberty of modernizing Smith’s use of pronouns.

[2]↩ As J. Richard Hackman taught us, the number of relationships (r) grows as a function number of members (n) per this equation: r = (n^2-n) / 2. And, each relationship bares some amount of coordination costs. See: Hackman, J. Richard. Leading teams: setting the stage for great performances. Boston, Mass.: Harvard Business School Press, 2002. Print.

[3]↩ It’s important to note that I am not suggesting that hiring full-stack data scientists results in fewer people overall. Rather, I am merely suggesting that when organized differently, their incentives are better aligned with learning vs. efficiency gains. Consider the following contrasting deptarment/team structures, each with 3 people. Fractional estimates and summed team sizes are illustrative only.

Specialist Model: organized for functional efficiency. Workers are not dedicated to any one business capability, rather their time is allocated to many.

Business Capability (columns)
Role (rows)
Recommendation System Algorithmic Inventory Management System Algorithmic Client Engagement System Team Size
ML Engineering 1/3 1/3 1/3 1
Modeling 1/3 1/3 1/3 1
Data Engineering 1/3 1/3 1/3 1
Total Dept Size 3

Generalists Model: Full-stack Data Scientists optimized for learning. Workers are fully dedicated to a business capability and perform all the functions.

Business Capability (columns)
Role (rows)
Recommendation System Algorithmic Inventory Management System Algorithmic Client Engagement System Total Dept Size
ML Engineering 1 1 1
Modeling
Data Engineering
Team Size 1 1 1 3

[4]↩ A more efficient way to learn about this approach to organization vs the trial and error I went through is to read the book by Amy C. Edmondson called “Teaming: How Organizations Learn, Innovate, and Compete in the Knowledge Economy” (Jossey-Bass, 2014).

[5]↩ This process of iteration assumes low cost of trial and error. If the cost of error is high you may want to rethink (i.e., it is not advised for medical or manufacturing). In addition, data volume and system availability requirements should also be considered. If you are dealing with petabytes or exabytes of data, specialization in data engineering may be warranted. Similarly, system availability (ie. uptime) and innovation are tradeoffs. If availability is paramount, functional excellence may trump learning. Finally, the full-stack data science model relies on the assumption of great people. They are not unicorns; they can be found as well as made. But they are in high demand and it will require certain conditions in order to attract and retain them (competitive compensation, company values, interesting work, etc.). Be sure your company culture can support this.

[6]↩ Smith, Adam. An inquiry into the nature and causes of the wealth of nations. Dublin: Printed for Messrs. Whitestone, 1776. Print. Page 464.

[7]↩ Pink, Daniel H.. Drive: the surprising truth about what motivates us. New York, NY: Riverhead Books, 2009.


          tf-nightly 1.14.1.dev20190313      Cache   Translate Page      
TensorFlow is an open source machine learning framework for everyone.
          tf-nightly-2.0-preview 2.0.0.dev20190313      Cache   Translate Page      
TensorFlow is an open source machine learning framework for everyone.
          Constructing mass-decorrelated hadronic decay taggers in ATLAS      Cache   Translate Page      
A large number of physics processes as seen by ATLAS at the LHC manifest as collimated, hadronic sprays of particles known as ‘jets’. Jets originating from the hadronic decay of a massive particle are commonly used in searches for both measurements of the Standard Model and searches for new physics. The ATLAS experiment has employed machine learning discriminants to the challenging task of identifying the origin of a given jet, but such multivariate classifiers exhibit strong non-linear correlations with the invariant mass of the jet, complicating many analyses which wish to make use of the mass spectrum. Adversarially trained neural networks (ANN) are presented as a way to construct mass-decorrelated jet classifiers by jointly training two networks in a domain-adversarial fashion. The use of neural networks further allows this method to benefit from high-performance computing platforms for fast development. A comprehensive study of different mass-decorrelation techniques is performed in ATLAS simulated datasets, comparing ANNs to designed decorrelated taggers (DDT), fixed-efficiency k-NN regression, convolved substructure (CSS), and adaptive boosting for uniform efficiency (uBoost). Performance is evaluated using metrics for background jet rejection and mass-decorrelation.
          Hardware Accelerated ATLAS Workloads on the WLCG      Cache   Translate Page      
In recent years the usage of machine learning techniques within data-intensive sciences in general and high-energy physics in particular has rapidly increased, in part due to the availability of large datasets on which such algorithms can be trained as well as suitable hardware, such as graphics or tensor processing units which greatly accelerate the training and execution of such algorithms. Within the HEP domain, the development of these techniques has so far relied on resources external to the primary computing infrastructure of the WLCG. In this paper we present an integration of hardware-accelerated workloads into the Grid through the declaration of dedicated queues with access to hardware accelerators and the use of linux container images holding a modern data science software stack. A frequent use-case of in the development of machine learning algorithms is the optimization of neural networks through the tuning of their hyper parameters. For this often a large range of network variations must be trained and compared, which for some optimization schemes can be performed in parallel -- a workload well suited for grid computing. An example of such a hyper-parameter scan on Grid resources for the case of Flavor Tagging within ATLAS is presented.
          [New report] Leveraging Data To Deliver Tailored Service Experiences      Cache   Translate Page      
The following research provides an overview of how companies are leveraging new and unusual forms of data to deliver more tailored, personalized and predictive services. From facial recognition and biometric sensors to computer vision and machine learning, this report provides innovative examples that showcase how companies are using more affective and cognitive data, collected by tracking and analyzing customers’ emotions, behaviors and preferences, to design high-touch product and service experiences that meet today’s elevated consumer expectations. Published March 2019 #adidas #ai #alexa #alibaba #amazon #app #apparel #artifical-intelligence #assisted-service #athos #automotive #beauty #biometrics #bmw #cardiogram #crm #customer-experience #customer-service #data #eloquii #fabletics #facial-recognition #fashion #fitness #health #intel-2 #john-hancock #levis #machine-learning #merchandising #merchandising-curation #mobile #new-balance #nike #north-face #papers #personalization #pinterest #post-purchase-service-support #predictive-services #premium #prose #rd #rent-the-runway #salesforce #sensoria #service-based-experiences #shopper-education #shopper-education-assistance #stitch-fix #store-experience #store-experience-design #strava #tmall #tommy-hilfiger #wellness
          Microsoft Dynamics 365 Engineer - Concurrency - Brookfield, WI      Cache   Translate Page      
Combining two decades of CRM and ERP developments with innovative technology such as machine learning and automation, Dynamics 365 is helping businesses around...
From Concurrency - Wed, 26 Dec 2018 20:30:31 GMT - View all Brookfield, WI jobs
          Data Architect - Data Warehouse & MPP - Nationwide Opportunities - Amazon Web Services, Inc. - Madison, WI      Cache   Translate Page      
These professional services engagements will focus on customer solutions such as Machine Learning, IoT, HPC, Batch/Real-time Data processing, MPP systems, Data...
From Amazon.com - Sat, 02 Feb 2019 10:11:05 GMT - View all Madison, WI jobs
          Sr Data Architect - Data Warehouse & MPP - Nationwide Opportunities - Amazon Web Services, Inc. - Madison, WI      Cache   Translate Page      
These professional services engagements will focus on customer solutions such as Machine Learning, IoT, HPC, Batch/Real-time Data processing, MPP systems, Data...
From Amazon.com - Wed, 30 Jan 2019 09:26:35 GMT - View all Madison, WI jobs
          Principal, Science & Analytics - CoreLogic - Milwaukee, WI      Cache   Translate Page      
Property Intel &amp; Risk Mgmt Sol. Perform pattern recognition model creation and training using various types of algorithms and machine learning modeling...
From CoreLogic - Thu, 07 Mar 2019 23:56:33 GMT - View all Milwaukee, WI jobs
          Sr. Business Intel. Engineer - Amazon.com Services, Inc. - Seattle, WA      Cache   Translate Page      
Use machine learning, data mining and statistical techniques to create new, scalable solutions for business problems....
From Amazon.com - Tue, 08 Jan 2019 21:02:56 GMT - View all Seattle, WA jobs
          Principal, Science & Analytics-Consumer Credit and Risk Modeling - CoreLogic - Irving, TX      Cache   Translate Page      
Strong business acumen. Regularly interact with internal and external experts in machine learning, economics, software programming, program management, and...
From CoreLogic - Mon, 25 Feb 2019 23:55:49 GMT - View all Irving, TX jobs
          Manufacturing General Manager - DataRobot - Columbus, OH      Cache   Translate Page      
DataRobot is one of the hottest AI start-ups in the world and Manufacturing has unlimited potential to take advantage of machine learning to transform their...
From DataRobot - Thu, 13 Dec 2018 12:36:01 GMT - View all Columbus, OH jobs
          Life Sciences General Manager - DataRobot - Columbus, OH      Cache   Translate Page      
DataRobot is one of the hottest AI start-ups in the world and Life Sciences has unlimited potential to take advantage of machine learning to transform their...
From DataRobot - Thu, 13 Dec 2018 12:36:01 GMT - View all Columbus, OH jobs
          Machine learning training in Bangalore (Bangalore)      Cache   Translate Page      
Machine learning gives computer the ability to learn from data by itself. This Machine learning training in Bangalore at NearLearn, explains how python community has developed many useful structures to achieve this. To Understand Machine Learning It requires a combination of multidisciplinary skills starting from an intersection of linear algebra, programming, statistics, computer science, calculus and business. Many researchers and programmers also think Machine learning is one of the best ways to make pr...
          "Nuestros periféricos no lo son de ordenadores y de dispositivos, lo son de la nube": Bracken Darrell, CEO de Logitech      Cache   Translate Page      

Todos reconoceréis a Logitech. Es una empresa que lleva con nosotros desde 1981, y prácticamente ha crecido paralelamente junto con el mercado de los ordenadores personales. Pero no lo ha hecho siguiendo un camino recto, ni mucho menos: Logitech ha vivido una serie de cambios que han hecho que sea una compañía irreconocible comparándola con sus inicios.

Para poder comentar los motivos de esta evolución, además de la visión actual y de futuro de la compañía, hemos charlado un rato con su CEO Bracken Darrell aprovechando su visita a Barcelona y su ponencia en el evento paralelo 4 Years From Now del Mobile World Congress 2019.

33 años fabricando ratones, y sin vistas a dejar de hacerlo

Logitech Raton Antiguo

Hagamos un resumen de la historia de Logitech. Desde el comienzo de la compañía en 1981 hasta 1998 fueron conocidos como fabricantes de ratones (aunque su idea inicial fallida fue ser una empresa de software). En esa época el mercado del ordenador personal no hacía más que crecer, y tal y como nos comenta Bracken la idea de fabricar más periféricos para los PC vino sola.

Así, Logitech vivió una edad de oro en la que creció en porcentajes de dos dígitos interanualmente pasando a fabricar teclados, altavoces y cámaras web entre otros accesorios para la época. Pero entonces llegó 2007, Apple presentó su iPhone y todo cambió. El mercado del ordenador personal empezó a ralentizar su crecimiento para después empezar a decaer. Y con él, todas las divisiones de Logitech empezaron a hacer lo mismo.

Fue ahí donde Bracken estrenó su cargo de CEO en Logitech. Hubo un cambio de estrategia en la compañía, que pasó a valorar más el diseño de sus productos y contrató al diseñador responsable hasta ese entonces de los productos de Nokia Alastair Curtis. Logitech salió de los números rojos diversificando objetivos, y vendiendo también accesorios para los dispositivos móviles.

Bracken Darrell 4yfn Bracken Darrell en la charla que ha dado en el evento 4YFN de Barcelona.

De todos modos, Bracken no quiere que la gente reconozca los productos de Logitech por su diseño. "Un gran diseño no implica un estilo específico y reconocible, al menos para nuestra compañía. Para nosotros un gran diseño es una experiencia de uso óptima. Puede que se puedan reconocer algunos detalles, pero el objetivo primario de nuestra empresa es lograr esa experiencia de usuario óptima. Lograr que digan "wow, este producto es perfecto", y que luego se den cuenta de que es de Logitech".

Los cambios que Logitech ha experimentado tampoco significan que la compañía haya dejado de vender ratones, desde luego. Siguen siendo un periférico esencial, y Bracken no considera que las previsiones de que sus ventas dejen de crecer 2023 sean ajustadas. El ejecutivo admite que no tiene una bola de cristal y que no se atreve a hacer previsiones que vayan más allá de dos años, pero para él no hay nada que se haya convertido en una mejor alternativa que el teclado y el ratón a pesar del crecimiento de otros modos de comunicación con nuestros gadgets:

"Las interfaces por voz son estupendas, pero no son lo mismo que el teclado y el ratón. Quizás van a actuar como algo complementario, pero no superarán del todo al ratón tradicional. Así que nuestro plan es seguir innovando en esos dispositivos, creo que ese mercado irá bien".

Bracken tampoco cree que las tabletas vayan a hacer que el ordenador desaparezca de nuestras vidas, mencionando incluso mi postura encorvada frente al portátil y defendiendo que lo mejor es tener una postura saludable frente a un monitor de sobremesa adecuadamente colocado. Aunque desde luego, "todo lo aprendido con los teclados de ordenador puede aplicarse a teclados como los de las tabletas".

"Los eSports son una parte increíblemente importante de nuestro negocio"

Esports

Un sector en el que los teclados y ratones de Logitech tienen mucha presencia es en el gaming. "Nuestra división de productos para jugones es ahora más grande que las de teclados y ratones por separado", nos afirma. Esa división, llamada Logitech G, ha cogido toda la experiencia de la fabricación de ratones y teclados que la compañía tiene desde 1981 y lo ha aplicado. El CEO está seguro de que cierto sector de clientes ven a Logitech como una compañía de artículos gaming, e insiste en que su principal ventaja es tener la experiencia de hacer teclados y ratones desde hace más de 33 años.

Y del gaming, nos vamos a los eSports. A mi pregunta sobre si los deportes electrónicos son interesantes para Logitech y si la compañía quiere tener un papel en ellos, Bracken me responde con un dato: los ratones inalámbricos de Logitech G fueron los usados por el equipo que ganó el campeonato de Overwatch el año pasado.

"Somos los máximos responsables de que los ratones inalámbricos estén entrando de lleno en el mercado del gaming en este momento", comenta Bracken. "Los Sports son una parte increíblemente importante para nosotros y nuestro modelo de negocio". Darrell también afirma que Logitech G crece más rápido que competidores como Razer, aunque quiere dejar claro que le encanta que haya competencia porque al final el beneficio recae en todo el mundo.

Los periféricos para la nube y otras revoluciones que veremos

Logitech Brio

¿Qué otros productos que hace Logitech se han popularizado? Pues las webcams, que también llevan siendo fabricadas por la compañía desde hace años. El sector creció durante un tiempo pero se estancó, sin embargo ahora mismo el negocio de las cámaras web vuelve a crecer gracias a los que se dedican a retransmitir contenidos a buena calidad desde plataformas como Twitch.

¿Y en cuanto a la privacidad? Tras recordarle a Bracken sobre las imágenes de Zuckerberg tapando la webcam de su propio portátil, la polémica está servida. El ejecutivo nos promete que Logitech ha aplicado "todos los protocolos de seguridad y privacidad" imaginables en sus cámaras, y que es algo que es crucial para la compañía. De hecho defiende el hecho de que sus webcams sean externas, ya que la mejor forma de protegerse es simplemente desconectándolas del ordenador en las que se estén usando.

Un ángulo interesante que también nos ofrece Bracken es el de cómo ha cambiado el modo con el que utilizamos nuestros dispositivos, tanto que hace que el ejecutivo diga que sus accesorios sirven para la nube más que para esos dispositivos.

"Mira este iPhone, por ejemplo [coge mi teléfono]. Es como si fuera el suelo de una pista de baloncesto: los servicios, como una pelota, rebotan encima de él constantemente. Nuestras cámaras de vídeo retransmiten hacia una nube. Nuestros altavoces reproducen canciones y podcasts que están en la nube. Los dispositivos son bases para todas las nubes que utilizamos".

Bracken Darrell 4yfn 2

Para el futuro a largo plazo, Bracken espera ver una revolución fruto de la evolución del Machine Learning. El ejecutivo también espera ver muchos beneficios fruto de la revolución que está suponiendo democratizar la creatividad: "Instagram, Twitch, YouTube. Todo el mundo puede ahora ser un artista, un cómico, un presentador, un retransmisor, un cineasta, un actor... simplemente desde su silla. Eso es una revolución creativa".

Más allá de eso, con la llegada del 5G y del Internet de las Cosas, Bracken espera ver una revolución en absolutamente todos los objetos que utilizamos. "Hasta esta mesa que estamos usando ahora, todo será reinventado por las startups". Respecto a Logitech, a largo plazo, Bracken se imagina verla como una compañía multimarca global de la misma forma como son ahora mismo Unilever o Procter&Gamble. Muchas marcas, muchas categorías de productos, englobadas en una gran corporación.

Imagen | bjornmeansbear

recommendation.header

Gasolina, diésel, híbrido… Cómo acertar con la propulsión

Dmitry Glukhovsky, autor de la saga &#39;Metro&#39;: &quot;Umberto Eco y Borges no ganaron el Nobel porque se atrevieron a entretener al público&quot;

Jesús Cañadas, ganador del premio Kelvin a la mejor novela fantástica: “El fantástico ha colonizado todas las facetas de nuestra vida cultural. Salvo la literaria”

-
the.news "Nuestros periféricos no lo son de ordenadores y de dispositivos, lo son de la nube": Bracken Darrell, CEO de Logitech originally.published.in por Miguel López .


          Exacerbations in Chronic Obstructive Pulmonary Disease: Identification and Prediction Using a Digital Health System      Cache   Translate Page      
Background: Chronic obstructive pulmonary disease (COPD) is a progressive, chronic respiratory disease with a significant socioeconomic burden. Exacerbations, the sudden and sustained worsening of symptoms, can lead to hospitalization and reduce quality of life. Major limitations of previous telemonitoring interventions for COPD include low compliance, lack of consensus on what constitutes an exacerbation, limited numbers of patients, and short monitoring periods. We developed a telemonitoring system based on a digital health platform that was used to collect data from the 1-year EDGE (Self Management and Support Programme) COPD clinical trial aiming at daily monitoring in a heterogeneous group of patients with moderate to severe COPD. Objective: The objectives of the study were as follows: first, to develop a systematic and reproducible approach to exacerbation identification and to track the progression of patient condition during remote monitoring; and second, to develop a robust algorithm able to predict COPD exacerbation, based on vital signs acquired from a pulse oximeter. Methods: We used data from 110 patients, with a combined monitoring period of more than 35,000 days. We propose a finite-state machine–based approach for modeling COPD exacerbation to gain a deeper insight into COPD patient condition during home monitoring to take account of the time course of symptoms. A robust algorithm based on short-period trend analysis and logistic regression using vital signs derived from a pulse oximeter is also developed to predict exacerbations. Results: On the basis of 27,260 sessions recorded during the clinical trial (average usage of 5.3 times per week for 12 months), there were 361 exacerbation events. There was considerable variation in the length of exacerbation events, with a mean length of 8.8 days. The mean value of oxygen saturation was lower, and both the pulse rate and respiratory rate were higher before an impending exacerbation episode, compared with stable periods. On the basis of the classifier developed in this work, prediction of COPD exacerbation episodes with 60%-80% sensitivity will result in 68%-36% specificity. Conclusions: All 3 vital signs acquired from a pulse oximeter (pulse rate, oxygen saturation, and respiratory rate) are predictive of COPD exacerbation events, with oxygen saturation being the most predictive, followed by respiratory rate and pulse rate. Combination of these vital signs with a robust algorithm based on machine learning leads to further improvement in positive predictive accuracy. Trial Registration: International Standard Randomized Controlled Trial Number (ISRCTN): 40367841; http://www.isrctn.com/ISRCTN40367841 (Archived by WebCite at http://www.webcitation.org/6olpMWNpc)

Purchases: 0
          Detecting Hypoglycemia Incidents Reported in Patients’ Secure Messages: Using Cost-Sensitive Learning and Oversampling to Reduce Data Imbalance      Cache   Translate Page      
Background: Improper dosing of medications such as insulin can cause hypoglycemic episodes, which may lead to severe morbidity or even death. Although secure messaging was designed for exchanging nonurgent messages, patients sometimes report hypoglycemia events through secure messaging. Detecting these patient-reported adverse events may help alert clinical teams and enable early corrective actions to improve patient safety. Objective: We aimed to develop a natural language processing system, called HypoDetect (Hypoglycemia Detector), to automatically identify hypoglycemia incidents reported in patients’ secure messages. Methods: An expert in public health annotated 3000 secure message threads between patients with diabetes and US Department of Veterans Affairs clinical teams as containing patient-reported hypoglycemia incidents or not. A physician independently annotated 100 threads randomly selected from this dataset to determine interannotator agreement. We used this dataset to develop and evaluate HypoDetect. HypoDetect incorporates 3 machine learning algorithms widely used for text classification: linear support vector machines, random forest, and logistic regression. We explored different learning features, including new knowledge-driven features. Because only 114 (3.80%) messages were annotated as positive, we investigated cost-sensitive learning and oversampling methods to mitigate the challenge of imbalanced data. Results: The interannotator agreement was Cohen kappa=.976. Using cross-validation, logistic regression with cost-sensitive learning achieved the best performance (area under the receiver operating characteristic curve=0.954, sensitivity=0.693, specificity 0.974, F1 score=0.590). Cost-sensitive learning and the ensembled synthetic minority oversampling technique improved the sensitivity of the baseline systems substantially (by 0.123 to 0.728 absolute gains). Our results show that a variety of features contributed to the best performance of HypoDetect. Conclusions: Despite the challenge of data imbalance, HypoDetect achieved promising results for the task of detecting hypoglycemia incidents from secure messages. The system has a great potential to facilitate early detection and treatment of hypoglycemia.

Purchases: 0
          Grab Buka Teknologi Keamanan untuk Partner Bisnis      Cache   Translate Page      

 

Liputan6.com, Jakarta Startup decacorn pertama di Asia Tenggara, Grab, membuka teknologi keamanannya yang bernama Grab Defence kepada partner bisnisnya.

Informasi, Grab Defence merupakan teknologi deteksi dan pencegahan kecurangan terbaru untuk mitra Grab.

Dengan Grab Defence, para mitra bisnis Grab dapat memanfaatkan kemampuan data perusahaan yang teruji dalam mengurangi tindak kecurangan, guna memperkuat ekosistem teknologi dan arus transaksi.

Berdasarkan laporan Cybersource SEA Fraud Benchmark 2018, 1,6 persen pendapatan e-commerce di Asia Tenggara hilang akibat tindak kecurangan. Khusus Indonesia, kerugian yang ditimbulkan akibat tindak kecurangan sebesar 3,2 persen.

Head of User Trust Grab, Wui Ngiap Foo, mengatakan, transaksi palsu dan manipulasi yang dilakukan oleh para pelaku kejahatan di berbagai platform, termasuk ride hailing, memberikan pengaruh besar pada ekonomi digital.

Oleh karenanya, selama beberapa tahun ini Grab berinvestasi besar untuk mengembangkan sistem yang lebih kuat dan didukung machine learning serta kecerdasan buatan untuk mengidentifikasi dan mencegah kecurangan pada platform Grab.

Hasilnya, menurut penelitian independen Spire Research and Consulting, tingkat penipuan di Grab kurang dari satu persen ketimbang kompetitor.

"Setiap hari, machine learning kami menganalisis jutaan data secara realtime untuk mendeteksi pola kecurangan, baik yang telah ada maupun baru. Tindak kecurangan terus berevolusi, oleh karenanya kami membangun algoritma yang berevolusi dan mempelajari polanya sehingga selangkah lebih maju dari kejahatan pelaku," kata Wui Ngiap Foo di Jakarta, Rabu (13/3/2019).

Menurutnya dengan algoritma yang realtime ini, kecurangan bisa dengan cepat ditindak.

Untuk itulah, Grab membuka teknologi keamanannya ini untuk para partner bisnisnya. "Kami ingin berbagi keahlian yang kami miliki dengan para mitra yang mungkin menghadapi masalah yang sama. Kita harus bahu membahu mengatasi masalah demi sistem teknologi yang lebih kuat di Asia Tenggara," tutur Wui Ngiap Foo.

 

Mitra Bisnis Bisa Akses Grab Defence

Fakta Tentang Startup Pertama Level Decacorn Grab#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

Wui Ngiap Foo mengatakan, Grab Defence yang bisa dipergunakan oleh mitra terdiri dari tiga hal, yakni event risk management suite yang memungkinkan pelaku bisnis bisa menilai risiko dari peristiwa atau transaksi.

"Ini terdiri dari serangkaian API untuk mengevaluasi risiko yang didukung oleh machine learning yang dipakai mitra bisnis untuk memprediksi risiko secara realtime, seperti investigasi atau analisis perilaku mencurigakan," katanya.

Kedua adalah entity intelligence services, yakni penggunaan database Grab serta keahlian mengidentifikasi berbagai jenis entitas perilaku kejahatan, misalnya nomor telepon, alamat email, dan lain-lain.

Contohnya, mitra bisnis bisa memakai database Grab untuk mendapatkan informasi nilai risiko dari pengguna baru. Jika angka risikonya rendah, mereka bisa mengizinkan pengguna masuk aplikasi tanpa lewat langkah tambahan.

Kemudian yang ketiga adalah device and network intelligence services. Dengan ini, pebisnis mampu mendeteksi pelaku kejahatan menggunkan data dari perangkat pengguna.

"Setiap bisnis yang melakukan transaksi online diuntungkan dengan Grab Defence. Pasalnya, Grab telah mendeteksi miliaran transaksi tiap tahunnya, sehingga mitra bisnis bisa memakai database dan teknologi Grab Defence untuk membantu keamanan bisnisnya," kata dia.

 

Kepercayaan Usur Penting dalam Platform Digital

Ekspansi GrabFood berkembang pesat baik di Indonesia maupun di seluruh Asia Tenggara (Instagram @grabid).#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

Sementara itu, Presiden Grab Indonesia, Ridzki Kramadibrata mengungkapkan, kepercayaan merupakan unsur yang sangat penting dalam platform digital.

"Sebagai everyday super app yang memiliki berbagai layanan kepercayaan merupakan hal utama. Salah satu unsur kepercayaan adalah keamanan. Grab didirikan pertama kali misinya adalah keamanan dan kenyamanan. Kami memiliki program yang keamanannya kuat dan kita melihat platform lain milik mitra juga penting untuk menerapkan keamanan," katanya.

Ridzki mengatakan, saat ini partner bisnis seperti OVO dan Kudo sudah bisa menggunakan Grab Defence, sementara parner Grab lainnya akan bisa menggunakan Grab Defence pada akhir 2019.

(Tin/Ysl)

Saksikan Video Pilihan Berikut Ini:


          Head of Marketing - 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 to make the...
From Wade & Wendy - Thu, 07 Mar 2019 17:00:19 GMT - View all New York, NY jobs
          Growth and Marketing Intern - 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, 18 Feb 2019 16:21:43 GMT - View all New York, NY jobs
          Product Operations Associate - 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 - Fri, 25 Jan 2019 16:22:23 GMT - View all New York, NY jobs
          AI Conversation Designer - 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 - Fri, 18 Jan 2019 08:17:41 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 - Fri, 18 Jan 2019 08:17:36 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 - Wed, 09 Jan 2019 08:24:46 GMT - View all New York, NY jobs
          Enterprise Marketing 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 to make the...
From Wade & Wendy - Thu, 22 Nov 2018 00:25:27 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
          Grab Buka Teknologi Keamanan untuk Partner Bisnis      Cache   Translate Page      

 

Liputan6.com, Jakarta Startup decacorn pertama di Asia Tenggara, Grab, membuka teknologi keamanannya yang bernama Grab Defence kepada partner bisnisnya.

Informasi, Grab Defence merupakan teknologi deteksi dan pencegahan kecurangan terbaru untuk mitra Grab.

Dengan Grab Defence, para mitra bisnis Grab dapat memanfaatkan kemampuan data perusahaan yang teruji dalam mengurangi tindak kecurangan, guna memperkuat ekosistem teknologi dan arus transaksi.

Berdasarkan laporan Cybersource SEA Fraud Benchmark 2018, 1,6 persen pendapatan e-commerce di Asia Tenggara hilang akibat tindak kecurangan. Khusus Indonesia, kerugian yang ditimbulkan akibat tindak kecurangan sebesar 3,2 persen.

Head of User Trust Grab, Wui Ngiap Foo, mengatakan, transaksi palsu dan manipulasi yang dilakukan oleh para pelaku kejahatan di berbagai platform, termasuk ride hailing, memberikan pengaruh besar pada ekonomi digital.

Oleh karenanya, selama beberapa tahun ini Grab berinvestasi besar untuk mengembangkan sistem yang lebih kuat dan didukung machine learning serta kecerdasan buatan untuk mengidentifikasi dan mencegah kecurangan pada platform Grab.

Hasilnya, menurut penelitian independen Spire Research and Consulting, tingkat penipuan di Grab kurang dari satu persen ketimbang kompetitor.

"Setiap hari, machine learning kami menganalisis jutaan data secara realtime untuk mendeteksi pola kecurangan, baik yang telah ada maupun baru. Tindak kecurangan terus berevolusi, oleh karenanya kami membangun algoritma yang berevolusi dan mempelajari polanya sehingga selangkah lebih maju dari kejahatan pelaku," kata Wui Ngiap Foo di Jakarta, Rabu (13/3/2019).

Menurutnya dengan algoritma yang realtime ini, kecurangan bisa dengan cepat ditindak.

Untuk itulah, Grab membuka teknologi keamanannya ini untuk para partner bisnisnya. "Kami ingin berbagi keahlian yang kami miliki dengan para mitra yang mungkin menghadapi masalah yang sama. Kita harus bahu membahu mengatasi masalah demi sistem teknologi yang lebih kuat di Asia Tenggara," tutur Wui Ngiap Foo.

 

Mitra Bisnis Bisa Akses Grab Defence

Fakta Tentang Startup Pertama Level Decacorn Grab#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

Wui Ngiap Foo mengatakan, Grab Defence yang bisa dipergunakan oleh mitra terdiri dari tiga hal, yakni event risk management suite yang memungkinkan pelaku bisnis bisa menilai risiko dari peristiwa atau transaksi.

"Ini terdiri dari serangkaian API untuk mengevaluasi risiko yang didukung oleh machine learning yang dipakai mitra bisnis untuk memprediksi risiko secara realtime, seperti investigasi atau analisis perilaku mencurigakan," katanya.

Kedua adalah entity intelligence services, yakni penggunaan database Grab serta keahlian mengidentifikasi berbagai jenis entitas perilaku kejahatan, misalnya nomor telepon, alamat email, dan lain-lain.

Contohnya, mitra bisnis bisa memakai database Grab untuk mendapatkan informasi nilai risiko dari pengguna baru. Jika angka risikonya rendah, mereka bisa mengizinkan pengguna masuk aplikasi tanpa lewat langkah tambahan.

Kemudian yang ketiga adalah device and network intelligence services. Dengan ini, pebisnis mampu mendeteksi pelaku kejahatan menggunkan data dari perangkat pengguna.

"Setiap bisnis yang melakukan transaksi online diuntungkan dengan Grab Defence. Pasalnya, Grab telah mendeteksi miliaran transaksi tiap tahunnya, sehingga mitra bisnis bisa memakai database dan teknologi Grab Defence untuk membantu keamanan bisnisnya," kata dia.

 

Kepercayaan Usur Penting dalam Platform Digital

Ekspansi GrabFood berkembang pesat baik di Indonesia maupun di seluruh Asia Tenggara (Instagram @grabid).#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

Sementara itu, Presiden Grab Indonesia, Ridzki Kramadibrata mengungkapkan, kepercayaan merupakan unsur yang sangat penting dalam platform digital.

"Sebagai everyday super app yang memiliki berbagai layanan kepercayaan merupakan hal utama. Salah satu unsur kepercayaan adalah keamanan. Grab didirikan pertama kali misinya adalah keamanan dan kenyamanan. Kami memiliki program yang keamanannya kuat dan kita melihat platform lain milik mitra juga penting untuk menerapkan keamanan," katanya.

Ridzki mengatakan, saat ini partner bisnis seperti OVO dan Kudo sudah bisa menggunakan Grab Defence, sementara parner Grab lainnya akan bisa menggunakan Grab Defence pada akhir 2019.

(Tin/Ysl)

Saksikan Video Pilihan Berikut Ini:


          Guavus Plus SQLstream means Broad and Deep for IoT Data Science      Cache   Translate Page      

History

From the first time that Damian Black, founder of SQLstream, and Dr. Anukool Lakhina, founder of Guavus, first met almost a decade ago, the synergies and complementary nature of their visions was apparent to both of them. Guavus was acquired by Thales in 2017. Thales, a large, international player in aerospace and defense, with a large presence in transportation, expressed interest in SQLstream about four years ago. It was at this point that Damian and Anukool realized that the solutions Guavus and SQLstream had developed since their earlier discussions, had become even more strongly complementary, with Guavus' deep domain expertise in telecommunications, machine learning and data science, and SQLstream as a pioneer and leader in streaming analytics with an horizontal platform. In addition, Guavus is following Thales lead in broadening their domain expertise into the Industrial Internet of Things. SQLstream has had great success in the Transportation area, and in other sensor analytics ecosystems (SensAE). In addition, Guavus recognizes the need to process the vast amount of telecoms and IoT data closer to the source.

Integration

Although the merger is only a month old, the two companies are already working together to bring the strengths of each together for greater customer success. Over the next six to 12 months, the two will be integrated into a single platform with the ability to scale up to mind-numbingly large data flows, and to very finely-tuned small aggregates where and as needed throughout the ecosystem. This will allow greater operational efficiency as separating signal from noise, close to the source, allows processing the data immediately, providing value timely and cost effectively. Data rates are growing, per Damian, by 50% as edge sources increase in importance, but data storage and management costs are only decreasing by 12-14%. Only by pushing the algorithms, the machine learning models, into the streaming pipeline, and Guavus has some of the best in the industry, customers in Telecom, Transportation, and IIoT in general, will organizations be able to actually draw value from this data.

With our integrated solutions, CSPs to IIoT customers will be able to take advantage of something that’s radically different as we deliver AI-powered analytics from the network edge to the network core. With this solution, our customers can now analyze their operational, customer, and business data anywhere in the network in real time, without manual intervention, so they can make better decisions, provide smarter new services, and reduce their costs." — Guavus Press Release

This matches well with what we have seen, and what we present in SensAE, that the ebb and flow of data throughout the ecosystem must allow for appropriate aggregation and analytics at each point within the ecosystem.

Future

At MWC19, there has been a lot of interest in these specific solution, and also in building trust throughout the ecosystem, with security, and with the ability to select the desirable levels of privacy and transparency. Responding to these industry concerns is already in the Thales/Guavus/SQLstream roadmap.

The SQLstream products have the ability to analyze, filter, and aggregate data at the network edge in real-time and forward the information to the network core where the Guavus’ Reflex® platform can apply AI-powered analytics, giving customers a widely distributed and scalable architecture with better price/performance and total cost of ownership." — Guavus Press Release

The next few months are going to be exciting with SQLstream, Guavus and Thales bringing together their expertise in streaming analytics, data management, telecommunications, transportation, machine learning, data science, industrial needs and system engineering.


          Python Machine Learning Blueprints, 2nd Edition epub      Cache   Translate Page      
none
          Hugh Howey - 2017 - Machine Learning New and Collected Stories (Sci-Fi)      Cache   Translate Page      
none
          Sr. DevOps Engineer - Machine Learning - CenturyLink - Phoenix, AZ      Cache   Translate Page      
Work closely with the Network and Customer Transformation organization individually or in collaborative teams to identify opportunities for leveraging company...
From CenturyLink - Sat, 02 Mar 2019 02:12:13 GMT - View all Phoenix, AZ jobs
          Principal Security Data Scientist - Sierra Nevada Corporation - Sparks, NV      Cache   Translate Page      
DATA SCIENCE / MACHINE LEARNING SKILLS:. Sierra Nevada Corporation is an Equal Opportunity Employer. Required to act as a trusted adviser for business leaders...
From Sierra Nevada Corporation - Fri, 15 Feb 2019 23:07:32 GMT - View all Sparks, NV jobs
          Data Engineer (Data Warehouse) - Bandwidth - Raleigh, NC      Cache   Translate Page      
Machine Learning &amp; statistics experience. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Mon, 25 Feb 2019 20:43:17 GMT - View all Raleigh, NC jobs
          Senior Software Engineer, Network Engineering - Bandwidth - Raleigh, NC      Cache   Translate Page      
You've got experience with machine learning and anomaly detection. We need developers who learn quickly and are comfortable diving into the operational...
From Bandwidth - Tue, 19 Feb 2019 16:35:34 GMT - View all Raleigh, NC jobs
          Database Admin Intern - Bandwidth - Raleigh, NC      Cache   Translate Page      
Familiarity with Machine Learning &amp; Statistical concepts. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Tue, 20 Nov 2018 22:34:57 GMT - View all Raleigh, NC jobs
          Data Engineer      Cache   Translate Page      
CA-Santa Clara, If you are a Data Engineer with experience, please read on! Top Reasons to Work with Us - Join a team who inspire each other to work towards greater innovative heights - We have the most secure, high performance, network services solution with built-in machine learning based analytics - You will work with incredibly smart, talented individuals and inspiring leaders with an outstanding track record
          Computer Scientist, GS-1550-09/11 (DEU-DP) - US Department of the Interior - Middleton, WI      Cache   Translate Page      
Researches existing machine learning techniques, and applies appropriate methods towards prediction. For additional information on our internal telework policy,... $50,598 - $79,586 a year
From usajobs.gov - Mon, 04 Mar 2019 10:08:38 GMT - View all Middleton, WI jobs
          Dynamics CRM Developer - Concurrency - Brookfield, WI      Cache   Translate Page      
Combining two decades of CRM and ERP developments with innovative technology such as machine learning and automation, Dynamics 365 is helping businesses around...
From Concurrency - Wed, 30 Jan 2019 20:34:33 GMT - View all Brookfield, WI jobs
          Microsoft Dynamics 365 Engineer - Concurrency - Brookfield, WI      Cache   Translate Page      
Combining two decades of CRM and ERP developments with innovative technology such as machine learning and automation, Dynamics 365 is helping businesses around...
From Concurrency - Wed, 26 Dec 2018 20:30:31 GMT - View all Brookfield, WI jobs
          Senior Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sat, 05 Jan 2019 10:09:26 GMT - View all New York, NY jobs
          Machine Learning Engineer - Temboo - New York, NY      Cache   Translate Page      
You will lead the design, prototyping and productization of machine learning-based features, and take responsibility for introducing other Temboo developers to...
From Temboo - Thu, 29 Nov 2018 10:09:26 GMT - View all New York, NY jobs
          Engineering Manager - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sun, 28 Oct 2018 10:09:38 GMT - View all New York, NY jobs
          Frontend Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role building web-based features on top of transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:31 GMT - View all New York, NY jobs
          Embedded Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:30 GMT - View all New York, NY jobs
          Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Fri, 28 Sep 2018 10:09:40 GMT - View all New York, NY jobs
          Transcription service Otter launches enterprise app for teams      Cache   Translate Page      
The Otter app uses AI and machine learning to render accurate transcripts that identify speakers, suggest keywords, and allow keyword searching.

          Lead Software Engineer - Java - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Streaming....
From Capital One - Thu, 28 Feb 2019 18:12:30 GMT - View all Tysons Corner, VA jobs
          Principal Associate Software Engineer - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), API Gateway, AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Kafka,...
From Capital One - Sat, 02 Feb 2019 15:41:05 GMT - View all Tysons Corner, VA jobs
          AI Platform Software Architect - DISCO - Austin, TX      Cache   Translate Page      
Should have prior experience designing, implementing, and operating scalable distributed systems for Big Data, Machine Learning or Machine Learning Problems....
From Disco - Tue, 12 Mar 2019 16:09:59 GMT - View all Austin, TX 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 - Thu, 20 Dec 2018 07:48:32 GMT - View all Providence, RI jobs
           Siber güvenlikte makine öğrenimi dönemine girdik      Cache   Translate Page      
Önümüzdeki dönemde siber saldırılarda otomasyon daha üst seviyeye çıkacak. Siber suçlular, veri toplama girişimlerinde yapay zeka (AI-Artificial Intelligence) sistemlerini ve makine öğrenimi (ML-Machine Learning) uygulamalarını artık daha fazla kullanacak. Böylece daha kişiselleştirilmiş ve sofistike kimlik avı saldırıları yapabilecekler.
          Machine Learning / Atificial Intelligence Engineer - Eigen - Grass Valley, CA      Cache   Translate Page      
Worldwide, Eigen has deployed two medical devices Artemis a 3D semi-robotic prostate biopsy system and ProFuse an MRI image fusion product....
From Indeed - Mon, 10 Dec 2018 18:58:13 GMT - View all Grass Valley, CA jobs
           Comment on Algorithms, Machine Learning, and Optimization: we are hiring! by güvenilir rulet siteleri       Cache   Translate Page      
nice share thanks
          Google uses its AI expertise to help the blind explore their surroundings      Cache   Translate Page      
Most of what Google has been doing in the AI space for the past couple of years is aimed at making life easier for people who already have it pretty easy. But the search giant occasionally uses its extensive knowledge of things like artificial intelligence and machine learning for a greater good, helping those less fortunate to leverage modern technology in a truly life-changing way.

Enter Lookout, a Google-designed mobile app targeted at the "nearly 253 million people in the world who are blind or visually impaired." Unfortunately, a very small percentage of that staggering ...
          Grab Defense, Cara Pesaing Gojek Tangkal Aksi Curang      Cache   Translate Page      
Grab Defense melibatkan teknologi machine learning.
          Tune: Control the comments you see | Jigsaw      Cache   Translate Page      
Also see Alphabet's AI-powered Chrome extension hides toxic comments | Engadget
"To test the idea of viewership control, today we are releasing an experimental Chrome extension called Tune that lets users customize how much toxicity they want to see in comments across the internet. Tune builds on the same machine learning models that power Perspective to let people set the “volume” of conversations on a number of popular platforms, including YouTube, Facebook, Twitter, Reddit, and Disqus. We hope Tune inspires developers to find new ways to put more control into the hands of readers to adjust the level of toxicity they see across the internet.
Tune lets you turn the volume of toxic comments down for “zen mode” to skip comments completely, or turn it up to see everything — even the mean stuff. Or you can set the volume somewhere in between to customize the level of toxicity (e.g. attacks, insults, profanity, etc) you’re willing to see in comments."
Tune: Control the comments you see | Jigsaw

          Offer - PRACTICAL SQL 2017 Video Training @ job support - FINLAND      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. SQL SERVER DBA Video Training Course Details: Real time training on SQL Server 2016 & 2017 DB Design and T-SQL. This training course is exclusively designed addressing all practical aspects of SQL Server fundamentals required for SQL DBA and Business Intelligence (MSBI) implementations. Material provided during the course. All Sessions are Completely Practical and Realtime. For free SQL Server DBA Video Demo, please visit : http://sqlschool.com/SQLDBA-Video-Training.html Schedules for PRACTICAL SQL 2016 & 2017 DBA Video TRAINING : http://sqlschool.com/Register.html Contact us today (24 x 7) for SQL DBA Practical Video Training SQL School Training Institute ISO 9001:2008 Certified Organization for Training Authorized Microsoft Partner (ID# 5108842) India: Mobile: +91 (0) 9666 44 0801 Mobile: +91 (0) 9666 64 0801 USA: Office: +1 (510) 400-4845 Office 1: #101, UMA Residency, Opp: Sindhu Travels, Beside Metro Station Gate #D, SR Nagar, Hyderabad - 38, India. . Website: http://sqlschool.com/ Follow us: https://www.facebook.com/sequelschool https://www.linkedin.com/company/sql-school https://twitter.com/sequelschool
          Technical Architect - AWS - CDW - Seattle, WA      Cache   Translate Page      
DevOps, Big Data, Machine Learning, Serverless computing etc. Solicit input/feedback from both internal and external customers to shape the service offering....
From CDW - Tue, 19 Feb 2019 18:49:04 GMT - View all Seattle, WA jobs
          Quality Assurance Engineer - Amazon.com Services, Inc. - Seattle, WA      Cache   Translate Page      
Do you want to use Amazon’s massive data sets and Machine Learning to do it? Our QAE will be able to understand software internals, debug problems using log...
From Amazon.com - Fri, 01 Feb 2019 10:10:42 GMT - View all Seattle, WA jobs
          Principal Associate Software Engineer - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), API Gateway, AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Kafka,...
From Capital One - Sat, 02 Feb 2019 15:41:05 GMT - View all Tysons Corner, VA jobs
          Technical Architect - AWS - CDW - Dallas, TX      Cache   Translate Page      
DevOps, Big Data, Machine Learning, Serverless computing etc. Solicit input/feedback from both internal and external customers to shape the service offering....
From CDW - Tue, 19 Feb 2019 18:49:04 GMT - View all Dallas, TX jobs
          Technical Architect - AWS - CDW - Austin, TX      Cache   Translate Page      
DevOps, Big Data, Machine Learning, Serverless computing etc. Solicit input/feedback from both internal and external customers to shape the service offering....
From CDW - Tue, 19 Feb 2019 18:49:04 GMT - View all Austin, TX jobs
          Senior Machine Learning Developer - CoStar Group - Richmond, VA      Cache   Translate Page      
CoStar Group is an Equal Employment Opportunity Employer; Senior Machine Learning Developer....
From CoStar Group - Fri, 22 Feb 2019 00:07:05 GMT - View all Richmond, VA jobs
          Sevilla acoge la 1ª edición de la Escuela de Machine Learning      Cache   Translate Page      
Durante la inauguración y clausura de la primera edición de la Escuela de Machine Learning en Sevilla, estuvieron presentes los secretarios generales de Empresa, Innovación y Emprendimiento y de Economía de la Junta de Andalucía, Manuel Ortigosa y Manuel Alejandro Hidalgo; el secretario general de Industria y de la Pyme y presidente de EOI, Raül Blanco; la directora general de EOI, Nieves Olivera; y el director de EOI Andalucía, Francisco Velasco
          Starszy Specjalista IT - Chatbot Development Team - Ing Bank Śląski - Katowice, śląskie      Cache   Translate Page      
Twoje zadania: Zaawansowana analiza danych z wykorzystaniem technik machine learning Tworzenie i utrzymywanie narzędzi usprawniających proces raportowania i...
Od Ing Bank Śląski - Tue, 12 Mar 2019 12:16:08 GMT - Pokaż wszystkie Katowice, śląskie oferty pracy
          Machine Learning / Atificial Intelligence Engineer - Eigen - Grass Valley, CA      Cache   Translate Page      
Worldwide, Eigen has deployed two medical devices Artemis a 3D semi-robotic prostate biopsy system and ProFuse an MRI image fusion product....
From Indeed - Mon, 10 Dec 2018 18:58:13 GMT - View all Grass Valley, CA jobs
          Data Scientist - Oliver Wyman - New York, NY      Cache   Translate Page      
Demonstrate solid and battle-tested understanding of the standard canon of machine learning practices, including but not limited to:....
From Marsh & McLennan Companies - Sat, 05 Jan 2019 15:04:15 GMT - View all New York, NY jobs
          Senior Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sat, 05 Jan 2019 10:09:26 GMT - View all New York, NY jobs
          Machine Learning Engineer - Temboo - New York, NY      Cache   Translate Page      
You will lead the design, prototyping and productization of machine learning-based features, and take responsibility for introducing other Temboo developers to...
From Temboo - Thu, 29 Nov 2018 10:09:26 GMT - View all New York, NY jobs
          Engineering Manager - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sun, 28 Oct 2018 10:09:38 GMT - View all New York, NY jobs
          Frontend Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role building web-based features on top of transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:31 GMT - View all New York, NY jobs
          Embedded Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:30 GMT - View all New York, NY jobs
          Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Fri, 28 Sep 2018 10:09:40 GMT - View all New York, NY jobs
          (IT) Senior Python Quant Developer      Cache   Translate Page      

Location: Zürich   

For our client, a major Swiss Bank, we are looking for a Senior Python Quant Developer a 24 months contract in Zurich. Senior Python Quant Developer Sector: Banking Duration: 24th months Location: Zurich, Switzerland Key Responsibilities A challenging role as a Senior Python Quant Developer in a demanding business (Risk Management and Portfolio Optimization). Duties and responsibilities include: Translating high level business requirements into quantitative models Implementation of requirements, with particular focus on scalability, stability, automated testing Interaction with an international team of developers, architects, quantitative investment professionals Essentials Skills and Qualifications Strong experience and knowledge of Python (+ 5 years) Knowledge of portfolio optimization techniques and theory, including principles of risk management in finance Understanding of financial time series and products, ability to understand and solve data quality issues in financial products Master studies in technical discipline from renowned university (math, computer science engineering, physics) concluded in rapid time and with excellent marks Previous experience of a development role in finance or computationally intense computing, eg banking, trading, asset management, risk management, insurance, scientific computing/academia Experience with Linux environments Team work Desired Skills and Qualifications: Experience with Python (and ideally with Java) Experience in DevOps tools (Docker, Kubernetes, cloud & containerization technologies) Previous experience with Agile methodologies Experience with data visualization and web GUIs (eg Flask) PhD in quantitative area recommended Interest in recent developments in AI & Machine Learning If this is the right opportunity for you, please apply online or contact Nadja Espey directly.
 
Type: Contract
Location: Zürich
Country: Switzerland
Contact: Nadja Espey
Advertiser: Harvey Nash IT Recruitment Switzerland
Reference: JS-354943/001

          Senior AI/Deep Learning Software Engineer - St Josephs Hospital and Medical Center - Phoenix, AZ      Cache   Translate Page      
Ability to align business needs to development and machine learning or artificial intelligence solutions. Experience in natural language understanding, computer...
From Dignity Health - Tue, 27 Nov 2018 03:06:49 GMT - View all Phoenix, AZ jobs
          Grow your games business with ads      Cache   Translate Page      

There’s so much that goes into building a great mobile game. Building a thriving business on top of it? That’s next level. Today, we’re announcing new solutions to increase the lifetime value of your players. Now, it’s easier than ever to re-engage your audience and take advantage of a new, smarter approach to monetization.

Help inactive players rediscover your game

Let's face it, the majority of players you acquire aren't going to continue engaging with your game after just a handful of days. One of the biggest opportunities you have to grow your business is to get those inactive players to come back and play again.

We’re introducing App campaigns for engagement in Google Ads to help players rediscover your game by engaging them with relevant ads across Google’s properties. With App campaigns for engagement, you can reconnect with players in many different ways, such as encouraging lapsed players to complete the tutorial, introducing new features that have been added since a player’s last session, or getting someone to open the game for the first time on Android (which only Google can help with).

Learn more about it here or talk to your Google account representative if you’re interested in trying it out.

Rediscover game 1

Generate revenue from non-spending players

Acquiring and retaining users is important, but retention alone doesn’t generate revenue.  Our internal data shows that, on average, less than four percent of players will ever spend on in-app items. One way to increase overall revenue is through ads. However, some developers worry that ads might hurt in-app purchase revenue by disrupting gameplay for players who do spend. What if you could just show ads to the players who aren't going to spend in your app? Good news—now you can.

We’re bringing a new approach to monetization that combines ads and in-app purchases in one automated solution. Available today, new smart segmentation features in Google AdMob use machine learning to segment your players based on their likelihood to spend on in-app purchases.

Ad units with smart segmentation will show ads only to users who are predicted not to spend on in-app purchases. Players who are predicted to spend will see no ads, and can simply continue playing.  Check it out by creating an interstitial ad unit with smart segmentation enabled.

Smart Segmentation Flow

To learn more about news ways to help you increase the lifetime value of your players, please join us at the Game Developers Conference. Location and details are below:


What: Google Ads Keynote
Where: Moscone West, room #2020
When: Wednesday March 20th at 12:30 PM


I'm excited for the week ahead and all the new games you’re building—I’m always on the lookout for my next favorite.



          (Senior) Engineer - Development and integration of Machine Learning products - SAP - Sankt Leon-Rot      Cache   Translate Page      
Requisition ID: 201307 Work Area: Software-Design and Development Location: Walldorf/St. Leon-Rot Expected Travel: 0 - 10% Career Status: Professional...
Gefunden bei SAP - Tue, 05 Mar 2019 12:36:55 GMT - Zeige alle Sankt Leon-Rot Jobs
          Grow your games business with ads      Cache   Translate Page      

There’s so much that goes into building a great mobile game. Building a thriving business on top of it? That’s next level. Today, we’re announcing new solutions to increase the lifetime value of your players. Now, it’s easier than ever to re-engage your audience and take advantage of a new, smarter approach to monetization.

Help inactive players rediscover your game

Let's face it, the majority of players you acquire aren't going to continue engaging with your game after just a handful of days. One of the biggest opportunities you have to grow your business is to get those inactive players to come back and play again.

We’re introducing App campaigns for engagement in Google Ads to help players rediscover your game by engaging them with relevant ads across Google’s properties. With App campaigns for engagement, you can reconnect with players in many different ways, such as encouraging lapsed players to complete the tutorial, introducing new features that have been added since a player’s last session, or getting someone to open the game for the first time on Android (which only Google can help with).

Learn more about it here or talk to your Google account representative if you’re interested in trying it out.

Rediscover game 1

Generate revenue from non-spending players

Acquiring and retaining users is important, but retention alone doesn’t generate revenue.  Our internal data shows that, on average, less than four percent of players will ever spend on in-app items. One way to increase overall revenue is through ads. However, some developers worry that ads might hurt in-app purchase revenue by disrupting gameplay for players who do spend. What if you could just show ads to the players who aren't going to spend in your app? Good news—now you can.

We’re bringing a new approach to monetization that combines ads and in-app purchases in one automated solution. Available today, new smart segmentation features in Google AdMob use machine learning to segment your players based on their likelihood to spend on in-app purchases.

Ad units with smart segmentation will show ads only to users who are predicted not to spend on in-app purchases. Players who are predicted to spend will see no ads, and can simply continue playing.  Check it out by creating an interstitial ad unit with smart segmentation enabled.

Smart Segmentation Flow

To learn more about news ways to help you increase the lifetime value of your players, please join us at the Game Developers Conference. Location and details are below:


What: Google Ads Keynote
Where: Moscone West, room #2020
When: Wednesday March 20th at 12:30 PM


I'm excited for the week ahead and all the new games you’re building—I’m always on the lookout for my next favorite.



          BrandPost: Next-Gen Data Center Networking – Built for AI, Powered by AI      Cache   Translate Page      

Artificial Intelligence/Machine Learning (AI/ML) is creating new opportunities for businesses and consumers. Indeed, PwC, a major consultancy, expects AI to contribute over $15.7 trillion to the global economy by 2030 [1]. At the same time, within the data center, a move to micro-services-based software architectures, distributed storage, and Artificial Intelligence/Machine Learning (AI/ML) workloads are pushing east-west traffic to previously unseen heights.

Huawei’s Global Industry Vision (GIV) 2025 [2] predicts that the AI adoption rate will reach 86% by 2025. This will also have a big impact on the data center network, when AI/ML applications will drive the data center towards the AI era from the cloud era. Huawei is pleased to commission AvidThink, formerly SDxCentral research, to author a research brief covering the impact on data center networking by new workloads, including Artificial Intelligence (AI) and Machine Learning (ML). If you’re ready to get your hands on a copy of “Next-Gen Data Center Networking - Built for AI, Powered by AI,” you can download a free copy from here. The following are some highlights from the report, though we’d still recommend you download and read through the 9 pages of content.

To read this article in full, please click here


          Senior Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sat, 05 Jan 2019 10:09:26 GMT - View all New York, NY jobs
          Machine Learning Engineer - Temboo - New York, NY      Cache   Translate Page      
You will lead the design, prototyping and productization of machine learning-based features, and take responsibility for introducing other Temboo developers to...
From Temboo - Thu, 29 Nov 2018 10:09:26 GMT - View all New York, NY jobs
          Engineering Manager - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Sun, 28 Oct 2018 10:09:38 GMT - View all New York, NY jobs
          Frontend Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role building web-based features on top of transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:31 GMT - View all New York, NY jobs
          Embedded Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Wed, 17 Oct 2018 10:09:30 GMT - View all New York, NY jobs
          Engineer - Temboo - New York, NY      Cache   Translate Page      
You’ll have a high-impact role working with transformative technologies like IoT and Machine Learning....
From Temboo - Fri, 28 Sep 2018 10:09:40 GMT - View all New York, NY jobs
          SolarWinds Extends IT Operations Management Leadership to Include Intelligent Anomaly Detection      Cache   Translate Page      
SolarWinds , a leading provider of powerful and affordable IT management software, today announced the availability of machine learning-enabled... Read more at VMblog.com.
          Addressing Fraud with Machine Learning: How & Why      Cache   Translate Page      

For the financial services industry (as well as many others that deal with data security and other types of non-monetary fraud), anomaly detection is hands down the most important system in operation. Yet many organizations still use more traditional modeling for fraud or anomaly detection instead of making the shift to machine learning.


          C3 launches Integrated Development Studio for speed up AI, machine learning deployments      Cache   Translate Page      
C3's approach with its development studio is designed to abstract processes across multiple clouds in a low-code environment.

          (IT) Senior Python Quant Developer      Cache   Translate Page      

Location: Zurich   

On behalf of our client, a leading financial institution in Zurich, Swisslinx is seeking a highly motivated Senior Python Quant Developer for a contract position of 12 months based in Zurich. You would be joining the Risk Management and Portfolio Optimization team, of an established financial organization. Your Mission: You would be joining the Risk Management and Portfolio Optimization team Translating high level business requirements into quantitative models Implementation of requirements, with particular focus on scalability, stability, automated testing Interaction with an international team of developers, architects, quantitative investment professionals Your Background: + 5 years of experience and knowledge of Python Knowledge of portfolio optimization techniques and theory, including principles of risk management in finance Understanding of financial time series and products, ability to understand and solve data quality issues in financial products Master studies in technical discipline from renowned university (math, computer science engineering, physics) concluded in rapid time and with excellent marks Previous experience of a development role in finance or computationally intense computing, eg banking, trading, asset management, risk management, insurance, scientific computing/academia Experience with Linux environments Fluent written and oral English Preferable skills: Experience with Java Experience in DevOps tools (Docker, Kubernetes, cloud & containerization technologies) Previous experience with Agile methodologies Experience with data visualization and web GUIs (eg Flask) PhD in quantitative area recommended Interest in recent developments in AI & Machine Learning Are you an experience Python Quant Developer looking for a highly-responsible role within an international environment? Then please apply online with your CV, we look forward to your application! By applying for this position, I consent to the Swisslinx Group of companies: - storing my personal information (including name, contact details, Identification and CV information etc.) on their internal or external Servers for the purpose of informing me of potential employment opportunities - using my personal information or - supplying it to third parties upon express consent for the purpose of informing me of potential job opportunities - transferring where applicable my personal information to a country outside the EEA/EFTA I also hereby agree to the Swisslinx privacy policy HTTP//www swisslinx com/en/legal/privacy-policy and Terms of Use HTTP//www swisslinx com/en/legal/disclaimer
 
Type: Contract
Location: Zurich
Country: Switzerland
Contact: Please click Apply
Advertiser: Swisslinx
Reference: JS10815

          Defining Elements of Knowledge Management      Cache   Translate Page      
Good non-technical piece in Forbes on elements of knowledge management that are addressed by taxonomies and ontologies.

Taxonomies, Ontologies and Machine Learning: The Future of Knowledge Management By Kurt Cagle in Cognitive World

Taxonomies, Ontologies and Machine Learning: The Future of Knowledge Management
"Taxonomies classify, ontologies specify"   ... 


          Handling Noise in Modeling      Cache   Translate Page      
In the Statistics.com Blog.  Useful piece on handling the inevitable noise with models.  Directional rather than technical piece.

Handling the Noise - Boost It or Ignore It?

Posted on Mar 06, 2019 By: Peter Bruce

In most statistical modeling or machine learning prediction tasks, there will be cases that can be easily predicted based on their predictor values (signal), as well as cases where predictions are unclear (noise). Two statistical learning methods, boosting and ProfWeight, use those difficult cases in exactly opposite ways - boosting up-weights them, and ProfWeight down-weights them. ... "


          Support Tools Developer - Pure Storage - Lehi, UT      Cache   Translate Page      
The world is experiencing a revolution driven by next-generation technology like AI, machine learning, virtual reality, quantum computing, and self-driving cars...
From Pure Storage - Thu, 28 Feb 2019 00:41:54 GMT - View all Lehi, UT jobs
          Knowledge Manager - Pure Storage - Lehi, UT      Cache   Translate Page      
The world is experiencing a revolution driven by next-generation technology like AI, machine learning, virtual reality, quantum computing, and self-driving cars...
From Pure Storage - Fri, 22 Feb 2019 22:51:05 GMT - View all Lehi, UT jobs
          Technical Support Engineer I - 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, 23 Dec 2018 08:19:51 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 - Fri, 30 Nov 2018 08:21:49 GMT - View all Lehi, UT jobs
          Senior Data Scientist - Predictive Enterprise Group - Neudesic LLC - Philadelphia, PA      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 - Sat, 15 Dec 2018 21:58:12 GMT - View all Philadelphia, PA 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
          NSBE Intern- National Society of Black Engineers 2019 National Convention - Visa - Detroit, MI      Cache   Translate Page      
Strong research experiences and publication record in machine learning and/or data mining, advanced cryptography, systems security, blockchain and quantum...
From Visa - Wed, 09 Jan 2019 04:11:40 GMT - View all Detroit, MI jobs
          Sr. Security Analyst II - AbbVie - Lake County, IL      Cache   Translate Page      
Understanding of Machine Learning. Coordinate efforts among multiple business units during Response. Interpret and summarize technical information for...
From AbbVie - Thu, 10 Jan 2019 21:07:17 GMT - View all Lake County, IL jobs
          Cleveland Clinic creates Center for Clinical Artificial Intelligence      Cache   Translate Page      
Work on machine learning algorithms focused on reducing risk of readmission and predicting patient response to cancer treatments.
          Why are marketers kidding themselves that AI is about more than sales?      Cache   Translate Page      

There will be many uses for machine learning in the future, but right now marketers should focus on improving sales by continually tweaking communications.

The post Why are marketers kidding themselves that AI is about more than sales? appeared first on Marketing Week.


          Ce soir, Paris Machine Learning #5 season 6: Explainable AI, Unity Challenge, Ethical AI      Cache   Translate Page      


Tonight, we will be hosted and sponsored by CFM capital. Thank you to them. 

The schedule is as followd :
6:45 Doors open
7PM - 9PM Talks
9PM - 10PM Cocktail - Networking

As usual, there is NO waiting list or reserved seat First come first served (the room has 110 seats)

This meetup will be streamed see below:



The presentations:

introduction to CFM Capital, Eric Lebigot

Vincent-Pierre Berges, The Obstacle Tower A Generalization Challenge in Vision, Control, and Planning, https://unity3d.com

The rapid pace of research in Deep Reinforcement Learning has been driven by the presence of fast and challenging simulation environments. These environments often take the form of games; with tasks ranging from simple board games, to classic home console games, to modern strategy games. We propose a new benchmark called Obstacle Tower: a high visual fidelity, 3D, 3rd person, procedurally generated game environment. An agent in the Obstacle Tower must learn to solve both low-level control and high-level planning problems in tandem while learning from pixels and a sparse reward signal. Unlike other similar benchmarks such as the ALE, evaluation of agent performance in Obstacle Tower is based on an agent's ability to perform well on unseen instances of the environment.

$100K AI Contest
Obstacle Tower Challenge
https://www.youtube.com/watch?v=nvdZpJkT-ls

=

Manar Toumi, Leornardo Noleto, Interpretability, https://www.bleckwen.ai

Machine learning interpretability is becoming an integral part of the data scientist workflow and can no longer be an afterthought. This talk will explore the vibrant area of machine learning interpretability and explain how to understand black-box models. Thanks to an interpretability technique based on colitional game theory: SHAP.

====

Cloderic Mars, craft.ai, Explainable AI

When it comes to actually leverage AI in production and especially in an environment where it interacts with humans, auditability and trust are not optional. That's why Explainable AI becomes a new R&D space. This talks will show why and where explainability in AI is needed, what it actually means and compare some of the techniques that falls into this category.

Arnaud de Moissac, https://dcbrain.com, Impact AI
Impact AI is a think and Do tank that aims to deal with the ethical and societal challenges of AI. We develop an ethical framework for responsible use of Artificial Intelligence respecting principles easy to understand and apply at a large scale. This talk is about the Governance part of this tool box

Join the CompressiveSensing subreddit or the Google+ Community or the Facebook page and post there !
Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.

          Data Engineer (Data Warehouse) - Bandwidth - Raleigh, NC      Cache   Translate Page      
Machine Learning &amp; statistics experience. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Mon, 25 Feb 2019 20:43:17 GMT - View all Raleigh, NC jobs
          Senior Software Engineer, Network Engineering - Bandwidth - Raleigh, NC      Cache   Translate Page      
You've got experience with machine learning and anomaly detection. 90-minute fitness lunch with a paid gym membership with shuttle service available for...
From Bandwidth - Tue, 19 Feb 2019 16:35:34 GMT - View all Raleigh, NC jobs
          TensorFlow.js 1.0      Cache   Translate Page      

Rilasciato TensorFlow.js 1.0, prima major release della libreria JavaScript che consente di implementare modelli di machine learning da browser Web.

Leggi TensorFlow.js 1.0


          Do You Want an Intern Position at Autodesk      Cache   Translate Page      
At Autodesk, we view our interns as real world employees and start them doing inspiring research and meaningful work collaborating with our teams. Ultimately we want to develop the interns into employees and they want to return after they graduate. Many of our employees started as interns in all of our business divisions including Autodesk Research where our interns typically work on research projects such as design, robotics, AI, machine learning, generative design, additive manufacturing, future casting and story telling,...
          BUSINESS INTELLIGENCE ANALYST - TransUnion - Portland, OR      Cache   Translate Page      
Identify and explore a creative project that utilizes your strengths in machine learning, business process development, global statistical analysis, or data...
From TransUnion - Wed, 06 Feb 2019 04:09:51 GMT - View all Portland, OR jobs
          SR. RESEARCH AND CONSULTING 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 - Tue, 29 Jan 2019 04:50:07 GMT - View all Chicago, IL jobs
          SR. RESEARCH AND CONSULTING MANAGER - 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 - Tue, 27 Nov 2018 10:04:56 GMT - View all Chicago, IL jobs
          Be your own broker with Abode HQ’s real estate platform      Cache   Translate Page      
Kyle Stoner, founder and CEO of Abode Technologies, joins Scott in-studio at WGN to talk about how easy real estate should be. Abode uses machine learning to match home buyers with top real estate professionals such as realtors, bankers, and contractors. Listen to learn more! This episode is sponsored by Bank of America and MB Real Estate. Listen to Technori on Spotify now!
          TensorFlow.js 1.0      Cache   Translate Page      
Rilasciato TensorFlow.js 1.0, prima major release della libreria JavaScript che consente di implementare modelli di machine learning da browser Web. Leggi TensorFlow.js 1.0

I virus e le vulnerabilità non vengono più aggiornate. Visitate il nostro sito per altri contenuti di qualità.
          How IBM Learns From Machine Learning      Cache   Translate Page      
With almost 40% of its research and development budget targeting blue sky research, IBM tries to keep its place in the fierce competition for the next computing era
          Cristiano Ronaldo Tak Pernah Terganggu oleh Tekanan      Cache   Translate Page      

Liputan6.com, Jakarta - Penyerang Juventus, Cristiano Ronaldo menepati janjinya. Sebelum laga Juventus vs Atletico Madrid di laga kedua 16 besar Liga Champions, Selasa (12/3) atau Rabu dini hari WIB, dia berjanji akan mencetak hattrick.

Hasilnya, itu yang terjadi. Cristiano Ronaldo mencetak tiga gol sekaligus membawa Juventus lolos ke permpat final setelah mengalahkan Atletico 3-0. Total, Juventus menang agregat 3-2.

Bukan kebetulan bila pemain berusia 34 tahun ini secara konsisten menghasilkan performa dan penampilan besar di panggung termegah seperti Liga Champions, dan sebuah penelitian baru-baru ini membuktikan kekuatan mentalnya yang luar biasa dalam situasi dengan pressure tinggi.

Penelitian tersebut dilakukan oleh perusahaan analisis olahraga SciSports. Dalam penelitiannya itu membuktikan bahwa Ronaldo tidak seperti yang lain di dunia sepak bola ketika itu berkaita dengan sebuah situasi atau momen bertekanan tinggi.

SciSports bekerja sama dengan universitas riset KU Leuven untuk mempelajari seberapa besar tekanan menjadi faktor dalam momen-momen penting dalam pertandingan sepak bola. Mereka mengumpulkan data dari 7.000 menit permainan, menganalisis bagaimana level performa seseorang dipengaruhi oleh tekanan dalam permainan dan coba tebak siapa yang berada di posisi teratas?

Cristiano [Ronaldo.(3915577 "") muncul di posisi puncak.

Kebal Tekanan 

Penyerang Juventus Cristiano Ronaldo (kanan) diikuti rekan-rekan setimnya merayakan gol ke gawang Atletico Madrid pada leg kedua babak 16 besar Liga Champions di Allianz Stadium, Turin, Selasa (12/3). Juventus menang 3-0. (Marco BERTORELLO/AFP)#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

Faktanya, penyerang Juventus yang juga pemenang Ballon d'Or lima kali tersebut memang seperti kebal terhadap setiap tekanan, dan itu tak mengejutkan terutama bila melihat bagaimana level performanya tak berubah di setiap skenario yang mungkin terjadi di permainan sepak bola.

Ronaldo menunjukkan seberapa kuat mentalnya ketika menghadapi tekanan saat melawan Atletico Madrid di leg kedua babak 16 besar Liga Champions. Juventus dituntut menang untuk mengejar defisit dua gol atas kekalahan di leg pertama, Ronaldo pun menunjukkannya. Bukan hanya satu gol, Ronaldo mencetak tiga gol dalam kemenangan 3-0 dan membawa Bianconeri ke perempat final Liga Champions dengan agregat 3-2.

Kemudian, seberapa besar dan sering tekanan memengaruhi performa pemain?

Tekanan Pengaruhi Permainan 

Dalam penelitian itu dilanjutkan juga bahwa ada beberapa pemain yang performa mereka sangat terpengaruh karena adanya tekanan.

Berdasarkan penelitian, Neymar adalah pemain yang performanya sangat dipengaruhi oleh tekanan di dalam permainan, dan demikian pula dengan bintang Chelsea, Eden Hazard.

Berdasarkan penelitian dari Analis SciSports, Jan van Haaren, Neymar membuat keputusan-keputusan yang justru lebih buruk bila berada di bawah tekanan. Sementara Hazard juga dipengaruhi oleh tekanan, dan sering membuat keputusan dalam permainan yang buruk ketika tim dalam penguasaan bola.

Profesor Jesse Davis dari Departemen Ilmu Komputer di KU Leuven, berbicara tentang penelitian tersebut.

"Tekanan mental sudah dipelajari di olahraga seperti baseball dan basket, tapi di sepak bola, ini yang belum dipetakan," ujarnya seperti dilansir SportBible.

"Itulah kenapa kami mengembangkan sebuah model menggunakan machine learning untuk memperkirakan seberapa banyak tekanan mental yang dialami oleh pemain yang menguasai bola," tambahnya.

"Model tersebut menganalisis bagaimana kinerja pemain ini di bawah tekanan: keputusan apa yang dia buat, apakah tindakan yang dipilih dieksekusi dengan baik dan seberapa besar dampak atas tindakan yang dipilih itu terhadap hasil pertandingan?" tutupnya.

Sumber: Bola.net


          Medio millar de startups optan a participar en la V edición del programa Decelera Menorca      Cache   Translate Page      

Decelera es el primer programa de desaceleración de startups tecno- sostenibles del mundo que plantea una metodología diferencial y única desde 2015. La iniciativa está asentada sobre cuatro valores: desaceleración, comunidad viva, factor humano y sostenibilidad. En él, emprendedores internacionales escapan del ruido de las operaciones diarias para reflexionar e inspirarse sobre cómo escalar su negocio y crecer de forma sostenida, y en cómo conseguir la financiación necesaria que les permita crecer.

Casi medio millar de startups se han inscrito para participar en su quinta edición. Los emprendedores seleccionados, que se darán a conocer el 1 de mayo, tendrán la oportunidad de hacer un paréntesis, conectar e impulsar sus proyectos (con potenciales inversores) durante dos semanas en la isla. En este tiempo, trabajarán con líderes del ecosistema startupero como Lucas Carné (Privalia), Max Kelly (Techstarts de Londres), Rupert Barksfield (Pynko), Manel Adell (ex CEO de Desigual) y Martin Varsavsky (Jazztel, Prelude and Overture), entre otros.

El programa de 18 meses de duración arranca con un paréntesis de entre una y dos semanas de desconexión en un entorno natural de singular belleza esencial para conseguir la inspiración y creatividad (Menorca y México). En las primeras jornadas, los startup heroes identifican las claves sobre cómo generar valor a través del trabajo en equipo, la importancia de la sostenibilidad –que es inherente a la desaceleración– y el poder la red de contactos y la comunidad.

En la segunda fase, se enfocan en sus proyectos y equipos mediante encuentros relajados con los experience makers, además de realizando actividades off-campus. Durante los últimos días tienen lugar los encuentros individuales con potenciales inversores para levantar financiación.

Las 494 startups que han aplicado al programa de Decelera Menorca en la isla (un 44% más que en 2018) son muy diversas tanto en la fase de desarrollo en que se encuentran –casi el 50% están en fase semilla, pero hay series A y series B como en el sector en el que trabajan. Inteligencia Artificial (AI) y Machine Learning, Blockchain, Big Data, IoT, son los mayoritarios entre las startups candidatas de 2019. Las candidatas a formar parte de la comunidad Decelera proceden de 77 países. De las inscripciones registradas, el 63% de las solicitudes provienen de Europa; un 14,4% de Asia; un 10,1% de África; un 8,7% de América del Norte; un 3,4% de Latinoamérica y un 0,4% de Oceanía. Por países, España concentra el mayor número de aspirantes 88, el 17,8% del total, seguido de India con 42 startups y Reino Unido, con 41.

La lista de las 25 ‘startups héroes’ de 2019 se dará a conocer a principios del mes de mayo, después de que el comité de evaluación internacional, compuesto por el equipo ejecutivo de Decelera, miembros de nuestra comunidad de antiguas ediciones, y socios fundadores e inversores de alta experiencia como Martin Varsavsky, termine el análisis pormenorizado de las compañías y sus equipos.

El 95% de las startups interesadas en participar en Decelera Menorca afirman que sus proyectos tecnológicos pretenden hacer este mundo mejor y más sostenible. Casi en su totalidad, las startups candidatas declaran estar alineadas con uno o varios de los 17 Objetivos de Desarrollo Sostenible (ODS) de las Naciones Unidas. En concreto, la mayoría reconoce que las innovaciones tecnológicas que promueven son motor de crecimiento y desarrollo económico que impulsan mejores infraestructuras y nuevas industrias.

"La calidad de las candidatas que han aplicado este año ha evolucionado favorablemente con respecto años anteriores. Estamos muy satisfechos de ello. Creemos que este cambio se debe a nuestra apuesta por fortalecer y desarrollar el primer programa que se preocupa por los equipos y el factor humano en el ecosistema , y por nuestro reposicionamiento de marca con Decelera", apuntó Marcos Martín, CEO y co-fundador de la primera desaceleradora de startups del mundo.


          Machine Learning Chip Market Growing Technology Trends and Business Opportunities by 2027      Cache   Translate Page      
(EMAILWIRE.COM, March 13, 2019 ) It is considered as a method of data analysis that automates analytical model building. Machine learning has been present from decades but not been widely used due to lack of big data it requires for processing. Over the last...
          IBM uses machine learning to detect early Alzheimer’s in blood samples      Cache   Translate Page      
A buildup of the protein amyloid-beta in spinal fluid can predict Alzheimer's long before symptoms appear, but it's difficult to detect. IBM researchers have identified four proteins that can be found in blood and can predict amyloid-beta buildup with up to 77% accuracy.
          Data Elixir - Issue 224      Cache   Translate Page      

In the News

The AI-Art Gold Rush Is Here

The gold rush started last October when Christie's sold an algorithm generated print for $432,500. More recently, an AI artist had its own show at a gallery in Chelsea. There's definitely a lot of interest here but is AI art really all that interesting? This longread in the Atlantic explores this burgeoning industry with links to artwork so you can judge for yourself.

theatlantic.com

Insight

Why Data Science Teams Need Generalists, Not Specialists

A team of specialists works well in environments where the organization knows exactly what needs to be done and execution can be managed like an assembly line. This article by Eric Colson explores why that's rarely the case in data science and how specialization can get in the way.

hbr.org

Sponsored Link

Master of Management Analytics: Your degree for the world of data

Realize the promise of data analytics and find the opportunity in the numbers. The Master of Management Analytics from Smith School of Business is essential training to unleash the potential of data and generate competitive advantage.

qns.bz

Tools and Techniques

Viewing Matrices & Probability as Graphs

Nice post that starts by showing how every matrix is a graph. From there, it's a visual tour of matrix operations and probabilities. Great read!

math3ma.com

Why Model Explainability is The Next Data Science Superpower

In this excerpt from his model explainability course, Dan Becker outlines the types of things that the very best data scientists are able to discern about their models and why that information is useful. This post also sparked a worthwhile discussion on Hacker News.

towardsdatascience.com

Exploring Neural Networks with Activation Atlases

Great interactive article on the Distil site that introduces a new technique for visualizing how decision-making happens in a neural network. It's a long read but it's compelling all the way through.

distill.pub

Set Your Jupyter Notebook up Right with this Extension

By default, Jupyter Notebooks are unnamed, have no markdown cells, and no imports. Since people are notoriously bad at changing default settings, why not encourage better practices? This simple extension gently nudges you to create better notebooks.

towardsdatascience.com

Lessons learned building natural language processing systems in health care

Building NLP systems in a complex domain like health care is hard. Not only do these systems require broad domain knowledge, every sub-specialty and form of communication is fundamentally different. In this post, David Talby outlines common issues and the lessons he's learned over 7 years of building NLP systems in health care.

oreilly.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

Awesome Machine Learning Interpretability

This curated list of machine learning interpretability resources is definitely worthy of its "awesome" moniker. Includes a blueprint of use-cases, software examples, tutorials, packages, books, papers, etc.

github.com

Data Viz

Data Visualization Society Logo: Behind the scenes

"Logo design" may not sound interesting but this post describes the logo for the newly formed Data Visualization Society. The logo changes dynamically according to member skills and it's unlike any logo you've ever seen.

medium.com

Jobs & Careers

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

dataelixir.com

About

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


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


          Amy Webb on Artificial Intelligence, Humanity, and the Big Nine      Cache   Translate Page      

BigNineCover-193x300.jpg
Futurist and author Amy Webb talks about her book, The Big Nine, with EconTalk host Russ Roberts. Webb observes that artificial intelligence is currently evolving in a handful of companies in the United States and China. She worries that innovation in the United States may lead to social changes that we may not ultimately like; in China, innovation may end up serving the geopolitical goals of the Chinese government with some uncomfortable foreign policy implications. Webb’s book is a reminder that artificial intelligence does not evolve in a vacuum–research and progress takes place in an institutional context. This is a wide-ranging conversation about the implications and possible futures of a world where artificial intelligence is increasingly part of our lives.

This week's guest:

This week's focus:

Additional ideas and people mentioned in this podcast episode:

A few more readings and background resources:

A few more EconTalk podcast episodes:

TimePodcast Episode Highlights
0:33

Intro. [Recording date: February 12, 2019.]

Russ Roberts: My guest is futurist and author Amy Webb.... Her latest book is The Big Nine: How the Tech Titans and Their Thinking Machines Could Warp Humanity.... Your book is a warning about the challenges we face, that we're going to face dealing with the rise of artificial intelligence. What is special about the book, at least in my experience reading about AI [Artificial Intelligence] and worries about artificial intelligence is that it doesn't talk about AI in the abstract but actually recognizes the reality that AI is mostly being developed within very specific institutional settings in the United States and in China. So, let's start with what you call the Big Nine. Who are they?

Amy Webb: Sure. So, what's important to note is that when it comes to AI, there's a tremendous amount of misplaced optimism and fear. And so, as you rightly point out, we tend to think in the abstract. In reality, there are 9 big tech giants who overwhelmingly are funding the research--building the open-source frameworks, developing the tools and the methodologies, building the data sets, doing the tests, and deploying AI at scale. Six of those companies are in the United States--I call them the G-Mafia for short. They are Google, Microsoft, Amazon, Facebook, IBM [International Business Machines], and Apple. And the other three are collectively known as the BAT. And they are based in China. That's Baidu, Alibaba, and Tencent. Together, those Big Nine tech companies are building the future of AI. And as a result, our helping to make serious plans and determinations, um, for I would argue the future of humanity.

Russ Roberts: And, just out of curiosity: I don't think you say very much in the book at all about Europe. Is there anything happening in Europe, in terms of research?

Amy Webb: Sure. So, the--you know, there's plenty of happening in France. Certainly in Canada. Montreal is one of the global hubs for what's known as Deep Learning. So this is not to say that there's not pockets of development and research elsewhere in the world. And it also isn't to say that there aren't additional large companies that are helping to grow the ecosystem. Certainly Salesforce and Uber are both contributing. However, when we look at the large systems, and the ecosystems and everything that plugs into them, overwhelming these are the 9 companies that we ought to be paying attention to.

3:18

Russ Roberts: So, I want to start with China. I had an episode with Mike Munger on the sharing economy and what he calls in his book Tomorrow 3.0. And, in the course of that conversation, we joked about people getting rated on their social skills and that those would be made public--how nice people were to each other. And we had a nice laugh about that. And I mentioned that I didn't think that that was an ideal situation--that people would be incentivized that way to be good people: despite my general love of incentives, that made me uneasy. And in response to that episode, some people mentioned an episode of Black Mirror[?]--the video series--and also some things that were happening in China. And I thought, 'Yeh, yeh, yeh, whatever.' But, what's happening in China--it's hard to believe. But, tell us about it.

Amy Webb: Sure. And, let me give you a quick example of one manifestation of this trend, and then sort of set that in the broader cultural context. So, there's a province in China where a new sort of global system is being rolled out. And it is continually mining and refining the data of the citizens who live in that area. So, as an example, if you cross the street when there's a red light and you are not able to safely cross the street at that point--if you choose to anyway, as to jay-walk--cameras that are embedded with smart recognition technology will automatically not just recognize that there's a person in the intersection when there's not supposed to be, but will actually recognize that person by name. So they'll use facial recognition technology along with technologies that are capable of recognizing posture and gait. It will recognize who that person is. Their image will be displayed on a nearby digital--not bulletin board; what do you call those--digital billboard. Where their name and other personal information will be displayed. And it will also trigger a social media mention on a network called Weibo. Which is one of the predominant social networks in China. And that person, probably, some of their family members, some of their friends, but also their employer, will know that they have--they have infracted--they have caused an infraction. So, they've crossed the street when they weren't supposed to. And, in some cases, that person may be publicly told--publicly shamed--and publicly told to show up at a nearby police precinct. Now, this is sort of important because it tells us something about the future of recognition technology and data. Which is very much tethered to the future of artificial intelligence. Now, better known as the Social Credit Score, China has been experimenting with this for quite a while; and they are not just tracking people as they cross the street. They are also looking at other ways that people behave in society, and that ranges from whether or not bills are paid on time, to how people perform in their social circles, to disciplinary actions that may be taken at work or at school, to what people are searching on--you know, on the Internet. And the idea is to generate some kind of a metric to show people definitively how well they are fitting in to Chinese society as Chinese people. This probably sounds, to the people listening to the show, like a horrible, Twilight Zone episode--

Russ Roberts: It sounds like 1984, is what it sounds like to me. It's not like, 'I wonder if that's a good idea.' It's more like, 'Are you kidding me?'

Amy Webb: Yeah. And so like, when I first heard about this, my initial response was not abject horror. I was curious. I was very curious.

Russ Roberts: [?]

Amy Webb: But like, here's what made me curious: Why bother? I mean, China has 1.4 billion people. And if the idea is to deploy something like this at scale, that is a tremendous amount of data. And you have to stop and say to yourself, 'Well, what's the point?' So, this is where some cultural context comes into play. So, I used to live in China. And I also used to live in Japan. And, they are very different cultures, very different countries. One distinctive feature of China is a community-reporting mechanism that is sort of embedded into society. And going back many thousands of years--you know, China is an enormous--it's a huge piece of land. And you've got people living throughout it; in fact, they are so spread apart, you have, you know, significantly different dialects being spoken. So, one way to sort of maintain control over vast masses of people spread out geographically was to develop a culture--sort of a tattle-tale culture. And so, throughout villages, if you were doing something untoward or breaking some kind of local custom or rule, that would get reported--you would get reported. Sort of in a gossipy way. But, you would get reported; and ultimately that person that heard the information would report that on up to maybe a precinct or a feudal manager of some kind, who would then report that up to whoever was in charge of the village or town; and then you would get into some kind of actual trouble. This was a way of maintaining social control. And so if you talk to people in China today, a lot of people are aware of monitoring. What I find so interesting is that at the moment, the outcry that we see outside of China does not match the outcry that I have observed--or actually to the lack of outcry--that I have observed in China. Now, there's one other piece of this really important: This is that using AI in this way ties in to China's Belt and Road Initiative [BRI]. And you might have heard about the BRI. This is sort of a master plan--it's a long-term strategy that helps China optimize what used to be the previous Silk Road--trading route. But it's sort of built around infrastructure. What's interesting is that there's also a digital version of this--the sort of digital BRI--where China is partnering with a lot of countries that are in situations where social stability is not a guarantee. And so, they are starting to export this technology into societies and places where there isn't that cultural context in place. And so, you have to stop and wonder and ask yourself, 'What does it mean for 58 pilot countries to have in their hands a technology capable of mining and refining and learning about all of their citizens, and reporting any infractions on up to authorities?' You know, in places like the Philippines, where free speech right now is questionable, this kind of technology, which does not make sense to us as Americans, may make slightly more sense to people in China, becomes a dangerous weapon in the hands of an authoritative, an authoritarian regime elsewhere in the world.

11:14

Russ Roberts: It reminds me, when you talk about the tattle-tale culture--of course, the Soviets did the same thing. They encouraged people to inform on--telltale sounds like a child reporting an insult. It's a monitoring mechanism by which authoritarian governments keep people in line. And you talk about the lack of outcry. Well, one reason is, is that you are worried that your social score is going to be low. Outcrying is probably not a good idea.

Amy Webb: That's right. That's right.

Russ Roberts: You should mention also, which I got from your book, that: It's not just like it's awkward, it's kind of embarrassing, you have a low score. These scores are going to be--going to be used, or being used?--to deal with people get credit, whether they can travel? Is that correct?

Amy Webb: Right. So, again. It's China. So, we can't be 100% of the information that's coming out, because it's a controlled-information ecosystem. But from what we've been able to gather, in all of the research that I've done, you know--I would suggest that it's already being used. It's certainly being used against ethnic minorities like the Uighurs. But we've seen instances of scoring systems being used to make determinations about school that kids are able to get into. You know, kids who, through no fault of their own may have parents that have run afoul, you know, in some way, and earned demotions and demerits on their social credit scores. So, it would appear as though this is already starting to affect people in China. And, again, my job is to quantify and assess future risk. So, as I was doing all of this research, my mind immediately went to: What are the longer term downstream implications? I think some of them are pretty obvious. Right? Like, some people in China are going to wind up having a miserable life as a result of the social credit score--the social credit score as it grows and is more widely adopted to some extent could lead to better social harmony, I guess; but it also leads to, you know, quashing individual ideas and certain freedoms and expressions of individual thinking. But, the flip side of this is: If it's the case that China has BRI--and it's investing in countries around the world not just in infrastructure but in digital infrastructure like fiber and 5G and communications networks and small cells and all the different technologies, in addition to AI and data, isn't it plausible that some time in the near future, our future trade wars aren't just rhetoric but could wind up in a retaliatory situation where people who don't have a credit score can't participate in the Chinese economy? Or, businesses that don't have credit scores can't do, can't trade. Or countries that don't have--if we think about like a Triple A Bond rating, you know, what happens if this credit scoring system evolves and China does business with, only with countries that have a high-enough score? We could quite literally get locked out of part of the global economy. It seems far-fetched, but I would argue that the signals are present now that point to something that could look like that in the near future.

15:03

Russ Roberts: Well, this is going to be a pretty paranoid show--episode--of EconTalk. So, I'm okay with that kind of fear-mongering, because it strikes me as quite worrisome. And I think we have to be, as you hinted at, you have to be open-minded that maybe this will make a better Chinese society, as defined by them. You know, the Soviets wanted to create a new Soviet man--and woman. They failed. But now, with these tools maybe there will be a new Chinese man and woman who will be harmoniously living with their neighbor, never jaywalking, and never gossiping, and smiling more often. Who knows? But, it's not my first, default thought about how this is going to turn out. I think that--

Amy Webb: No, but you kind of--you have to start with--I want to point out that I am not like a dystopian fiction writer. I'm a pragmatist. So, this--I am not studying all of this for the purpose of scaring people. What I would argue is, I have studied all of this, and used data, and modeled out plausible outcomes; and it is scary. It really is. Because you have to, again, connect the dots between all of this and other adjacent areas that are important to note. The CCP [Chinese Communist Party] in China is--

Russ Roberts: the Communist Party--

Amy Webb: yep--is facing some huge opportunities but also big problems. The Chinese economy may technically be slowing, but it's not a slow economy. There's plenty of growth ahead. And, if that holds--and there's no reason why at the moment it wouldn't--you know, Chinese society is about to go through social mobility at a scale never seen before in modern human history. And as that enormous group of people moves up, they are going to want to buy stuff. They are going to want to travel. So, you know, that potentially causes some problems, because the more wealth that is earned, the more agency people feel, the more opinions they start having about how the government ought to be run. And, you know, the CCP effectively made the current President of China, Xi Jinping, effectively President for life. And 2049--which seems far off but in the grand scheme of things isn't really that far into the future--is the 100th anniversary of the founding of the CCP. China is very good at long-term planning. Now, they've not always made good on fulfilling promises. But they are good at planning.

Russ Roberts: Yes, they are.

Amy Webb: Right? So, I don't see all of this as flashes in the pan, and 'AI's kind of a hot buzzy topic right now.' I'm looking at the much longer term and the much bigger picture. That's what makes me kind of concerned.

18:02

Russ Roberts: I think that's absolutely right. One other institutional detail to make clear for listeners is that the Chinese Internet is roped off, to some extent--to quite a large extent. They are developing their own tools and apps. And, talk about the three companies in China that are working on AI and how they work together in a way that American companies are not.

Amy Webb: So, here's another interesting facet of the Big Nine and AI is on a sort of a dual-developmental track. In China, Baidu, Alibaba, and Tencent were all formed sort of in the late 1990s, early 2000s; and their origin stories are not all that different from our big, modern tech giants like Amazon and Google and Apple. The key distinction is that our big tech companies were formed out for the most part in Seattle, Redmond, and Cupertino--California and San Francisco. Where, the ecosystem was able to blossom: there's plenty of competition. And there was plenty of talent. California has fairly lenient--in some ways--fairly lenient employer/employee laws which has made it very easy for talent to move between companies. And, if you are somebody who studies innovation, you know, the sort of lack of--the limited or lack of regulation, the ability for people to move around--

Russ Roberts: letting people make enormous amounts of money when they succeed and losing all of it when they fail--

Amy Webb: Right. Right. Right. But, the lack of safety net, the lack of a central, federal authority, if you will, is partly what enabled these companies to grow. And to grow fast. And to grow big. Which is why we also see a lot of overlap. So, Google, Microsoft, Amazon, and IBM [International Business Machines] own and maintain the world's largest cloud infrastructure. So, if you own a website or you are a business owner or you are making a phone call, at some point you are accessing one of their clouds. You know--we have competing, for the most part, we have competing operating systems for our mobile devices. For the most part, we still have competing email systems. And that's because without a central authority dictating one of the companies was going to do which thing, they all sort of did it. They went alone. When it went on their own and built their own things. So, now we have tremendous wealth concentrated among just a few companies who own the lion's share of patents; who are funding most of the research. And, for the most part, Silicon Valley and Washington, D.C. have an antagonistic relationship. That is not the case in China. So, in China, when the big tech companies were being formed there, you don't do anything in China without also in some way creating that business in concert with Beijing--with the government. You've got to pull patents--I'm sorry--you've got to pull permits. You have to abide by various regulations and laws. People are checking in on you. So, while Baidu, Alibaba, and Tencent may be independent financial organizations, in practical terms they are very much working in lockstep with Beijing. Alibaba, for those of you not familiar with the company, is very similar to Amazon. So, it's a retail operation. Tencent is very similar to our social media: so, sort of Twitter meets gaming and chat. And, I'm sorry--and Baidu is sort of search--is the sort of Google-esque company of the bunch. When China--when the Chinese government decided that AI was going to be a central part of its future plans--and this was decided years ago--it also decided that Tencent was going to focus on health; that Baidu was going to focus on cloud; and that Alibaba was going to focus on various different data aspects. I'm sorry; and Baidu was also going to focus on AI and transportation. So, it's not as though these companies came to these additional areas of research and work on their own. It was centrally coordinated. And that's a really, really important thing to keep in mind. If we've got a central government, a powerful government that is now--that has this long-term vision and is centrally coordinating, what's happening at a top level with the research and the movements of these companies, suddenly you have a streamlined system where you don't have arguments about regulation; you don't have the companies at each other's throats--like we've seen in the United States, Apple suddenly calling for sweeping privacy regulations because, to be fair, it's sort of--they are already far ahead and it gives them a competitive advantage. You don't see all that infighting in China. So, we have some fundamental differences. And the real challenge is that while we're trying to sort all this out in the United States, you have a streamlined central authority with three very powerful companies who are all now collaborating in some way on the future. In addition to a bunch of other top-level government initiatives to repatriate academics; to bring back top AI people; but also to do things like start educating kindergarteners about AI. There is a textbook that is going to roll out this year throughout China teaching kindergarteners the fundamentals of machine learning. I mean, you know--whereas in the United States, you know, some of our government officials, you know, up until very recently denied AI's capabilities; and only yesterday--so this is February 11th--President Trump issued an Executive Order to, I guess--I mean, there's a handful of bullet points on what AI ought to be, but it wasn't a policy paper. There's no funding. There's no government structure set up. There's not--I mean it--you see where I'm getting at?

25:07

Russ Roberts: Well, yeah--let me push back against that a little bit. You know, China is growing tremendously; as you point out, they are going to, presumably, they are already in one of the greatest transformations in human history from the countryside to the city, from a low standard of living to a much higher standard of living. And most of that's wonderful, and I'm happy about it. We don't know exactly what their ambitions will be or are outside of their own borders, and therefore what the repercussions are for us. As you suggest, they are doing a bunch of stuff. But the fact that they are top down and planning and organized, and we are chaotic and disorganized--so, just to take an example, you know, there's n companies in America, more than 4; I don't know how many there are--working on various aspects of driverless autonomous vehicles. There's Uber; there's Lyft; Apple; Google; there's Waymo. There's a lot going on here. And a lot of that will turn out. That's the nature of creative destruction; and capitalism. Some of those investments won't pan out. It will--the gambles will fail and lose, and people lose all their money. And, in general, historically, that chaotic soup of competition serves the average person and the people who are innovators quite well. The fact that China has, say, Baidu focusing on that and no one else having to worry about it, could be a bug, not a feature. I'm not convinced that China teaching kindergarteners machine learning is going to turn out well. Could be a mistake. Could be an enormous blunder. They are not allowing kind of experimentation, trial and error, that in my view is central to innovation. So, I think it remains to be seen how successful their walled garden with top-down gardening going on from the government's vision of what they want AI to serve, is going to work out. It might. It could. And it could be hard--the outcomes might be really bad for not just the Chinese but for other people. But it might just kind of fail. And, I'm not even convinced that their growth path is going to continue the way it has in the past. A lot of people just assume that because they have grown dramatically over the last 25 years they'll keep growing dramatically. There's a lot of ghost cities in China; there's a lot of overbuilding. I'm not so sure they have everything under control. So, I think you have to have that caveat as a footnote to those concerns.

Amy Webb: I completely agree with you. I would say that, for years, especially in the United States, we've been indoctrinated into thinking that China is a copy-paste culture rather than a culture that understands how to innovate, and to some extent I think that that is the result of that heavy-fisted, top-down approach to business. What I'm concerned about is not whether China succeeds financially. Here's what I'm concerned about. The challenge with artificial intelligence is that it's already here. It is not--there's no event horizon. There's no single thing that happens. It's already here. And it's been here for a while. And, in fact, it powers--you know, artificial [?] intelligence now powers our email; it powers the anti-lock brakes in our cars. You know. And essentially, this new Third Era of computing that we are in, if we assume that the First Era was tabulation--so that would have been Ada Lovelace in the late 1800s--and a Second Era was programmable systems, which would have been those early IBM mainframes on up to the, you know, desktop computers that we use today. This next Era is AI. And AI, while we've seen it anthropomorphized in movies like Her and on shows like Westworld, at its heart, AI is simply systems that make decisions on our behalf. And they do that using tools to optimize. So, the challenge is that, right now, systems are capable of making fairly narrow decisions. And the structures of those systems, and which data they were trained on, and how they make decisions and under what circumstances, those decisions were made by a relatively few number of people working at the BAT [Baidu, Alibaba, Tencent] in China and at the G-Mafia here in the United States. And the problem is that these systems aren't static. They continue to learn. And they--you know--they join, literally millions and millions of other algorithms that are all working in service of optimizing things on our behalf. Which is why I agree with you that if we are talking about a self-driving future, it's good to have competition, because--for all the usual reasons. Right? We get better form factors[?]; we get better vehicles; we get better price points. But when we are talking about systems that are continuing to evolve, that grow more and more powerful the more data they have access to and the more compute they are given--more computer power. And as we move into the more technical aspects, there are things like Generative Adversarial Networks, which are specifically designed to play tricks, to help systems learn more quickly. We are talking about slowly but surely ceding control over to systems to make these decisions on our behalf. And, that is what concerns me. What concerns me is that we do not have a singular set of guardrails that are global in nature. We don't have norms and standards. I'm not in favor of regulation. On the other hand, we don't have any kind of agreed-upon ideas for who and what to optimize for, under what circumstances. Or even what data sets to use. And China has a vastly different approach than we do in the United States, in part because China has a completely different viewpoint on what details of people's private lives should be mined, refined, and productized. And here in the United States, a lot of these companies have obfuscated when and how they are using our data. And, the challenge is that we all have to live with the repercussions.

32:10

Russ Roberts: Yeah, I'd agree with that. Up to a point. I want to give you a chance to talk about some scary examples. I think the--I'll just say, up front, that for me, underlying this whole problem--there are many different proximate causes and concerns. But there is, it seems to me, a very significant lack of competition. We can talk about how much competition there is in the United States relative to China. But certainly--the concern for me here in the United States is that the Big Six[Big Nine?] here in the United States will stay the Big Six[Big Nine?]. Which will give them leverage to do a bunch of things that you or I might not like. I do want to add that whatever we do to regulate or constrain them, via culture or whatever, allows for the possibility that they don't stay the Big Six[Big Nine?]. And I think one of the challenges of any way to deal with these problems is that, if you're not careful, you are going to end up creating a cartel that--it's de facto right now, but that can change. But if you make it de jure, you're going to end up with much worse outcomes than I think we're going to have. But, to concede your point about concern: I do think the Silicon Valley ethos of ask for forgiveness rather than permission--because right now there's no one you have to ask permission for, generally. Users are not paying much attention. There's very little regulation of how your private data is being used. Obviously something happened on January 1st, 2019 because I get a lot of annoying bars on my websites saying 'Will you accept cookies?' and I stupidly always click 'Yes,' like I'm sure most people do. And now they've complied with whatever required them to do that, and they're moving along. So, you know, I do think that there are some serious issues here. And you give some examples in the book of where these corporations--or China--have done things, and they really pay a price for it. They just keep going. The Facebook/Cambridge Analytica problem. The example you give of China pressuring Marriott the way their website was designed in terms of territorial recognition of China's sovereignty over various places that are somewhat up in the air. Those are serious issues, I think. And, more importantly, they are just the tip of the iceberg. So, talk about a couple of those things that you are worried about, that I think are alarming. And, normally, the marketplace would punish these folks; but not much does.

Amy Webb: So, I love what you just said, which is that the market--so it's curious, right? Why has the marketplace not punished the Big Nine? Or at least the G-Mafia, right? Or at least Facebook?

Russ Roberts: They've been punished a little bit. I think their users are down. I'm thinking about deleting my Facebook page. And I'm sure--and I've switched to DuckDuckGo for my searching. It's a really small step. But these are things that maybe people are starting to do in a little, slightly bigger, numbers.

Amy Webb: Maybe. But, again, like I don't have access to the whole world's data. Thank God. But--and you--let's just reveal our biases: like, you and I are digitally savvy people.

Russ Roberts: You're kind, Amy.

Amy Webb: Well, but you are. I think the fact that you even know what DuckDuckGo is, that you are somebody who is using it, I think is quite telling. [More to come, 35:37]

(21 COMMENTS)

          IA : IBM développe des modèles de machine learning pour la détection précoce de la maladie d'Alzheimer à partir d'un simple test sanguin      Cache   Translate Page      
IA : IBM développe des modèles de machine learning pour la détection précoce de la maladie d'Alzheimer
à partir d'un simple test sanguin

S'il y a un domaine où l'intelligence artificielle peut être utilisée pour le bien de la société, c'est probablement dans le domaine médical où l'IA peut fournir une assistance précieuse aux spécialistes humains. L'IA est de plus en plus utilisée dans ce corps de métier pour aider les médecins dans le diagnostic des patients, surtout pour des cas non évidents qui...
          Lead Software Engineer - Java - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Streaming....
From Capital One - Thu, 28 Feb 2019 18:12:30 GMT - View all Tysons Corner, VA jobs
          Principal Associate Software Engineer - Capital One - Tysons Corner, VA      Cache   Translate Page      
Java Ecosystem (Spring Boot, Maven, Tomcat etc.), API Gateway, AWS (S3, ECS etc.), Angular, JUnit, Jasmine, Karma, Microservices, Machine Learning, Kafka,...
From Capital One - Sat, 02 Feb 2019 15:41:05 GMT - View all Tysons Corner, VA jobs
          AI Platform Software Architect - DISCO - Austin, TX      Cache   Translate Page      
Should have prior experience designing, implementing, and operating scalable distributed systems for Big Data, Machine Learning or Machine Learning Problems....
From Disco - Tue, 12 Mar 2019 16:09:59 GMT - View all Austin, TX 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 - Thu, 20 Dec 2018 07:48:32 GMT - View all Providence, RI jobs
          Machine learning tracks moving cells      Cache   Translate Page      
(Okinawa Institute of Science and Technology (OIST) Graduate University) Scientists can now study the migration of label-free cells at unprecedented resolution, a feat with applications across biology, disease research, and drug development.
          We need an algorithmic bill of rights before algorithms do us wrong      Cache   Translate Page      

As corporations rapidly roll out advanced artificial intelligence, we need to develop strong safeguards capable of meeting the risks.

Michael Kearns, a leading machine learning researcher, recently addressed a group of 60 esteemed scientists at the prestigious Santa Fe Institute. His subject was the insidious bias built into many of the algorithms used for socially sensitive activities such as criminal sentencing and loan approval.

Read Full Story


          Implementation Engineer - Behavox - New York, NY      Cache   Translate Page      
Knowledge and application of statistical methods and Machine Learning algorithms. The IM team is owning an increasing number of deployment and integration tasks...
From Behavox - Mon, 10 Dec 2018 16:19:54 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
          Applied Scientist - Machine Learning - Amazon.com Services, Inc. - Bellevue, WA      Cache   Translate Page      
The Last Mile Science &amp; Technology organization plays a crucial role in developing solutions to vehicle routing problems critical for Amazon Logistics, Prime...
From Amazon.com - Tue, 26 Feb 2019 07:51:35 GMT - View all Bellevue, WA jobs
          Product Manager - CRM Web Solutions, LLC - Bedford, TX      Cache   Translate Page      
Communicate feature releases to internal and external users. Experience in working on AI, Machine Learning products is a huge plus....
From CRM Web Solutions, LLC - Tue, 12 Mar 2019 15:31:15 GMT - View all Bedford, TX jobs
          How to Build Your Own Search Ranking Algorithm with Machine Learning by @CoperniX      Cache   Translate Page      

Here's how you can build your very own web search ranking algorithm using machine learning.

The post How to Build Your Own Search Ranking Algorithm with Machine Learning by @CoperniX appeared first on Search Engine Journal.


          L’intelligenza artificiale è l’elemento in più dei team aziendali      Cache   Translate Page      
Ai e machine learning sono destinati sempre più a rappresentare un’arma fondamentale per i chief information officer e per la vita di un’impresa. Ma è fondamentale che queste soluzioni tecnologiche partano da informazioni valide: i giusti dataset serviranno per migliorare a loro volta gli algoritmi. Il punto di vista di Avon Puri, Cio di Rubrik.
          PLC Systems Market Influenced by the Growing Demand for IoT and Machine learning in Industrial Sector      Cache   Translate Page      

IndustryARC offers PLC systems market research report which emphasizes on the recent developments as well as historical evidence. The report states that, as of 2018, the global PLC systems market size was approximately $6.0 to $6.5 billion, and its value will increase at a CAGR of 12 to 15% over the course of the forecast […]

The post PLC Systems Market Influenced by the Growing Demand for IoT and Machine learning in Industrial Sector appeared first on Red Newswire.


          Machine Learning Engineering Intern - EY - Palo Alto, CA      Cache   Translate Page      
AI Labs in Palo Alto, CA, under the leadership of the Global AI Leader, Nigel Duffy, is in the process of revolutionizing EY and the the auditing industry with...
From EY - Thu, 07 Mar 2019 01:39:45 GMT - View all Palo Alto, CA jobs
          Going ATOMIC: Clustering and Associating Attacker Activity at Scale « Going ATOMIC: Clustering and Associating Attacker Activity at Scale      Cache   Translate Page      

At FireEye, we work hard to detect, track, and stop attackers. As part of this work, we learn a great deal of information about how various attackers operate, including details about commonly used malware, infrastructure, delivery mechanisms, and other tools and techniques. This knowledge is built up over hundreds of investigations and thousands of hours of analysis each year. At the time of publication, we have 50 APT or FIN groups, each of which have distinct characteristics. We have also collected thousands of uncharacterized 'clusters' of related activity about which we have not yet made any formal attribution claims. While unattributed, these clusters are still useful in the sense that they allow us to group and track associated activity over time.

However, as the information we collect grows larger and larger, we realized we needed an algorithmic method to assist in analyzing this information at scale, to discover new potential overlaps and attributions. This blog post will outline the data we used to build the model, the algorithm we developed, and some of the challenges we hope to tackle in the future.

The Data

As we detect and uncover malicious activity, we group forensically-related artifacts into 'clusters'. These clusters indicate actions, infrastructure, and malware that are all part of an intrusion, campaign, or series of activities which have direct links. These are what we call our "UNC" or "uncategorized" groups. Over time, these clusters can grow, merge with other clusters, and potentially 'graduate' into named groups, such as APT33 or FIN7. This graduation occurs only when we understand enough about their operations in each phase of the attack lifecycle and have associated the activity with a state-aligned program or criminal operation.

For every group, we can generate a summary document that contains information broken out into sections such as infrastructure, malware files, communication methods, and other aspects. Figure 1 shows a fabricated example with the various 'topics' broken out. Within each 'topic' – such as 'Malware' – we have various 'terms', which have associated counts. These numbers indicate how often we have recorded a group using that 'term'.

image
Figure 1: Example group 'documents' demonstrating how data about groups is recorded

The Problem

Our end goal is always to merge a new group either into an existing group once the link can be proven, or to graduate it to its own group if we are confident it represents a new and distinct actor set. These clustering and attribution decisions have thus far been performed manually and require rigorous analysis and justification. However, as we collect increasingly more data about attacker activities, this manual analysis becomes a bottleneck. Clusters risk going unanalyzed, and potential associations and attributions could slip through the cracks. Thus, we now incorporate a machine learning-based model into our intelligence analysis to assist with discovery, analysis, and justification for these claims.

The model we developed began with the following goals:

  1. Create a single, interpretable similarity metric between groups
  2. Evaluate past analytical decisions
  3. Discover new potential matches

image
Figure 2: Example documents highlighting observed term overlaps between two groups

The Model

This model uses a document clustering approach, familiar in the data science realm and often explained in the context of grouping books or movies. Applying the approach to our structured documents about each group, we can evaluate similarities between groups at scale.

First, we decided to model each topic individually. This decision means that each topic will result in its own measure of similarity between groups, which will ultimately be aggregated to produce a holistic similarity measure.

Here is how we apply this to our documents.

Within each topic, every distinct term is transformed into a value using a method called term frequency -inverse document frequency, or TF-IDF. This transformation is applied to every unique term for every document + topic, and the basic intuition behind it is to:

  1. Increase importance of the term if it occurs often with the document.
  2. Decrease the importance of the term if it appears commonly across all documents.

This approach rewards distinctive terms such as custom malware families – which may appear for only a handful of groups – and down-weights common things such as 'spear-phishing', which appear for the vast majority of groups.

Figure 3 shows an example of TF-IDF being applied to a fictional "UNC599" for two terms: mal.sogu and mal.threebyte. These terms indicate the usages of SOGU and THREEBYTE within the 'malware' topic and thus we calculate their value within that topic using TF-IDF. The first (TF) value is how often those terms appeared as a fraction of all malware terms for the group. The second value (IDF) is the inverse of how frequently those terms appear across all groups. Additionally, we take the natural log of the IDF value, to smooth the effects of highly common terms – as you can see in the graph, when the value is close to 1 (very common terms), the log evaluates to near-zero, thus down-weighting the final TF x IDF value. Unique values have a much higher IDF, and thus result in higher values.

image
Figure 3: Breakdown of TF-IDF metric when evaluated for a single group in regard to malware

Once each term has been given a score, each group is now reflected as a collection of distinct topics, and each topic is a vector of scores for the terms it contains. Each vector can be conceived as an arrow, detailing the 'direction' that group is 'pointing'within that topic.

Within each topic space, we can then evaluate the similarity of various groups using another method – Cosine Similarity. If, like me, you did not love trigonometry – fear not! The intuition is simple. In essence, this is a measure of how parallel two vectors are. As seen in Figure 4, to evaluate two groups' usage of malware, we plot their malware vectors and see if they are pointing in the same direction. More parallel means they are more similar.

image
Figure 4: Simplified breakdown of Cosine Similarity metric when applied to two groups in the malware 'space'

One of the nice things about this approach is that large and small vectors are treated the same – thus, a new, relatively small UNC cluster pointing in the same direction as a well-documented APT group will still reflect a high level of similarity. This is one of the primary use cases we have, discovering new clusters of activity with high similarity to already established groups.

Using TF-IDF and Cosine Similarity, we can now calculate the topic-specific similarities for every group in our corpus of documents. The final step is to combine these topic similarities into a single, aggregate metric (Figure 5). This single metric allows us to quickly query our data for 'groups similar to X' or 'similarity between X and Y'. The question then becomes: What is the best way to build this final similarity?

image
Figure 5: Overall model flow diagram showing individual topic similarities and aggregation in to final similarity matrix

The simplest approach is to take an average, and at first that’s exactly what we did. However, as analysts, this approach did not sync well with analyst intuition. As analysts, we feel that some topics matter more than others. Malware and methodologies should be more important than say, server locations or target industries...right? However, when challenged to provide custom weightings for each topic, it was impossible to find an objective weighting system, free from analyst bias. Finally, we thought: "What if we used existing, known data to tell us what the right weights are?" In order to do that, we needed a lot of known – or "labeled" – examples of both similar and dissimilar groups.

Building a Labeled Dataset

At first our concept seemed straightforward: We would find a large dataset of labeled pairs, and then fit a regression model to accurately classify them. If successful, this model should give us the weights we wanted to discover.

Figure 6 shows some graphical intuition behind this approach. First, using a set of ‘labeled’ pairs, we fit a function which best predicts the data points.

image
Figure 6: Example Linear regression plot – in reality we used a Logistic Regression, but showing a linear model to demonstrate the intuition

Then, we use that same function to predict the aggregate similarity of un-labeled pairs (Figure 7).

image
Figure 7: Example of how we used the trained model to predict final similarity from individual topic similarities.

However, our data posed a unique problem in the sense that only a tiny fraction of all potential pairings had ever been analyzed. These analyses happened manually and sporadically, often the result of sudden new information from an investigation finally linking two groups together. As a labeled dataset, these pairs were woefully insufficient for any rigorous evaluation of the approach. We needed more labeled data.

Two of our data scientists suggested a clever approach: What if we created thousands of 'fake' clusters by randomly sampling from well-established APT groups? We could therefore label any two samples that came from the same group as definitely similar, and any two from separate groups as not similar (Figure 8). This gave us the ability to synthetically generate the labeled dataset we desperately needed. Then, using a linear regression model, we were able to elegantly solve this 'weighted average' problem rather than depend on subjective guesses.

image
Figure 8: Example similarity testing with 'fake' clusters derived from known APT groups

Additionally, these synthetically created clusters gave us a dataset upon which to test various iterations of the model. What if we remove a topic? What if we change the way we capture terms? Using a large labeled dataset, we can now benchmark and evaluate performance as we update and improve the model.

To evaluate the model, we observe several metrics:

  • Recall for synthetic clusters we know come from the same original group – how many do we get right/wrong? This evaluates the accuracy of a given approach.
  • For individual topics, the 'spread' between the calculated similarity of related and unrelated clusters. This helps us identify which topics help separate the classes best.
  • The accuracy of a trained regression model, as a proxy for the 'signal' between similar and dissimilar clusters, as represented by the topics. This can help us identify overfitting issues as well.

Operational Use

In our daily operations, this model serves to augment and assist our intelligence experts. Presenting objective similarities, it can challenge biases and introduce new lines of investigation not previously considered. When dealing with thousands of clusters and new ones added every day from analysts around the globe, even the most seasoned and aware intel analyst could be excused for missing a potential lead. However, our model is able to present probable merges and similarities to analysts on demand, and thus can assist them in discovery.

Upon deploying this to our systems in December 2018, we immediately found benefits. One example is outlined in this blog post about potentially destructive attacks. Since then we have been able to inform, discover, or justify dozens of other merges.

Future Work

Like all models, this one has its weaknesses and we are already working on improvements. There is label noise in the way we manually enter information from investigations. There is sometimes 'extraneous' data about attackers that is not (yet) represented in our documents. Most of all, we have not yet fully incorporated the 'time of activity' and instead rely on 'time of recording'. This introduces a lag in our representation, which makes time-based analysis difficult. What an attacker has done lately should likely mean more than what they did five years ago.

Taking this objective approach and building the model has not only improved our intel operations, but also highlighted data requirements for future modeling efforts. As we have seen in other domains, building a machine learning model on top of forensic data can quickly highlight potential improvements to data modeling, storage, and access. Further information on this model can also be viewed in this video, from a presentation at the 2018 CAMLIS conference.

We have thus far enjoyed taking this approach to augmenting our intelligence model and are excited about the potential paths forward. Most of all, we look forward to the modeling efforts that help us profile, attribute, and stop attackers.


          Manager, Data Science - 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 - Fri, 30 Nov 2018 00:47:34 GMT - View all Boise, ID jobs
          Sr. Manager, Media Analytics - Micron - Folsom, CA      Cache   Translate Page      
Statistics, probability theory, heuristics and machine learning. This means conducting business with integrity, accountability, and partnership while supporting...
From Micron - Thu, 21 Feb 2019 18:56:44 GMT - View all Folsom, CA jobs
          Director Venture Capital - 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, 08 Feb 2019 00:52:01 GMT - View all Milpitas, CA jobs
          Understanding Recruitment: Machine Learning Engineer (NLP, DNNs, Reinforcement Learning)      Cache   Translate Page      
£60000 - £65000 per annum + discounts: Understanding Recruitment: Machine Learning Engineer (NLP, DNNs, Reinforcement Learning) We are looking for a Machine Learning Engineer (NLP, DNNs, Reinforcement Learning) to join us and work on novel projects that involve fast system prototyping and experimentation, at a company h Hatfield
          Infosys to open innovation hub in Romania      Cache   Translate Page      
The company said the facility at southeastern European country would focus on developing offerings for clients based on digital technologies including cloud, big data, artificial intelligence and machine learning.
          Applied Scientist - Machine Learning - Amazon.com Services, Inc. - Bellevue, WA      Cache   Translate Page      
The Last Mile Science &amp; Technology organization plays a crucial role in developing solutions to vehicle routing problems critical for Amazon Logistics, Prime...
From Amazon.com - Tue, 26 Feb 2019 07:51:35 GMT - View all Bellevue, WA jobs
          Product Manager - CRM Web Solutions, LLC - Bedford, TX      Cache   Translate Page      
Communicate feature releases to internal and external users. Experience in working on AI, Machine Learning products is a huge plus....
From CRM Web Solutions, LLC - Tue, 12 Mar 2019 15:31:15 GMT - View all Bedford, TX jobs
          Machine Learning and Big Data Engineer - Makrwatch - Cali, Valle del Cauca      Cache   Translate Page      
We believe video will grow exponentially over the coming years and content creators will continue to challenge mass media as we know it.... $4.000.000 - $9.000.000 al mes
De Indeed - Fri, 15 Feb 2019 18:38:03 GMT - Ver todos: empleos en Cali, Valle del Cauca
          Senior AI/Deep Learning Software Engineer - St Josephs Hospital and Medical Center - Phoenix, AZ      Cache   Translate Page      
Ability to align business needs to development and machine learning or artificial intelligence solutions. Experience in natural language understanding, computer...
From Dignity Health - Tue, 27 Nov 2018 03:06:49 GMT - View all Phoenix, AZ jobs
          Dell EMC, Nvidia make AI reference architecture available      Cache   Translate Page      
AI is becoming a critical workload for enterprises and storage giants are rolling out building blocks to build out machine learning and AI workloads.

          PyTorch Geometric: A Fast PyTorch Library for DL      Cache   Translate Page      
A new GitHub project, PyTorch Geometric (PyG), is attracting attention across the machine learning community. PyG is a geometric deep learning extension library for PyTorch dedicated to processing irregularly structured input data such as graphs, point clouds, and manifolds.
          Supervisor Test Engineering (6)- 69876 - Advanced Micro Devices, Inc. - Austin, TX      Cache   Translate Page      
Follow tech trends and understand impacts to AMD business. Machine learning experience a plus. What you do at AMD changes everything....
From Advanced Micro Devices, Inc. - Mon, 10 Dec 2018 19:32:25 GMT - View all Austin, TX jobs
          Machine Learning - Accenture - Bengaluru, Karnataka      Cache   Translate Page      
Accenture Technology powers our clients’ businesses with innovative technologies—established and emerging—changing the way their people and customers experience...
From Accenture - Tue, 12 Mar 2019 15:46:15 GMT - View all Bengaluru, Karnataka jobs
          Comment on A Clear Definition of Machine Learning by Nya Jackson      Cache   Translate Page      
This was helpful. I often get confused with the difference between AI, machine learning and automation so I appreciate you explaining how they relate and differ.
          Data Scientist - Oliver Wyman - New York, NY      Cache   Translate Page      
Demonstrate solid and battle-tested understanding of the standard canon of machine learning practices, including but not limited to:....
From Marsh & McLennan Companies - Sat, 05 Jan 2019 15:04:15 GMT - View all New York, NY 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 - Tue, 25 Dec 2018 09:45:46 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 - Tue, 25 Dec 2018 09:45:46 GMT - View all Palo Alto, CA jobs
          Intel, Huawei y HPE, entre los promotores de estándar que evita cuellos de botella en centros de datos      Cache   Translate Page      

En muchos centros de datos hay ciertos problemas relacionados con el tráfico entre las CPUs y chips aceleradores. Hablamos de cuellos de botella, que frenan la expansión y el desarrollo de las cargas de trabajo, que crecen con rapidez debido a tecnologías como la Inteligencia Artificial y el Machine Learning. Para evitarlo, varias de las […]

La entrada Intel, Huawei y HPE, entre los promotores de estándar que evita cuellos de botella en centros de datos aparece primero en MuyComputerPRO.


          Apple acquires Laserlike machine learning startup      Cache   Translate Page      
Apple has acquired Laserlike, a small Silicon Valley-based machine learning startup...
          CryptoNumerics Announces Free Downloadable CN-Protect Software That Uses AI To Create Privacy Protected Datasets While Maintaining Their Quality For Machine Learning. CryptoNumerics Also Announces A $2.5 Million USD Seed Financing From Leading Silicon Valley Venture Capital Firms      Cache   Translate Page      


          Predicting Future Cardiovascular Events in Patients With Peripheral Artery Disease Using Electronic Health Record Data.      Cache   Translate Page      
Related Articles

Predicting Future Cardiovascular Events in Patients With Peripheral Artery Disease Using Electronic Health Record Data.

Circ Cardiovasc Qual Outcomes. 2019 Mar;12(3):e004741

Authors: Ross EG, Jung K, Dudley JT, Li L, Leeper NJ, Shah NH

Abstract
BACKGROUND: Patients with peripheral artery disease (PAD) are at risk of major adverse cardiac and cerebrovascular events. There are no readily available risk scores that can accurately identify which patients are most likely to sustain an event, making it difficult to identify those who might benefit from more aggressive intervention. Thus, we aimed to develop a novel predictive model-using machine learning methods on electronic health record data-to identify which PAD patients are most likely to develop major adverse cardiac and cerebrovascular events.
METHODS AND RESULTS: Data were derived from patients diagnosed with PAD at 2 tertiary care institutions. Predictive models were built using a common data model that allowed for utilization of both structured (coded) and unstructured (text) data. Only data from time of entry into the health system up to PAD diagnosis were used for modeling. Models were developed and tested using nested cross-validation. A total of 7686 patients were included in learning our predictive models. Utilizing almost 1000 variables, our best predictive model accurately determined which PAD patients would go on to develop major adverse cardiac and cerebrovascular events with an area under the curve of 0.81 (95% CI, 0.80-0.83).
CONCLUSIONS: Machine learning algorithms applied to data in the electronic health record can learn models that accurately identify PAD patients at risk of future major adverse cardiac and cerebrovascular events, highlighting the great potential of electronic health records to provide automated risk stratification for cardiovascular diseases. Common data models that can enable cross-institution research and technology development could potentially be an important aspect of widespread adoption of newer risk-stratification models.

PMID: 30857412 [PubMed - in process]


          Clinical Value of Predicting Individual Treatment Effects for Intensive Blood Pressure Therapy.      Cache   Translate Page      
Related Articles

Clinical Value of Predicting Individual Treatment Effects for Intensive Blood Pressure Therapy.

Circ Cardiovasc Qual Outcomes. 2019 Mar;12(3):e005010

Authors: Duan T, Rajpurkar P, Laird D, Ng AY, Basu S

Abstract
BACKGROUND: The absolute risk reduction (ARR) in cardiovascular events from therapy is generally assumed to be proportional to baseline risk-such that high-risk patients benefit most. Yet newer analyses have proposed using randomized trial data to develop models that estimate individual treatment effects. We tested 2 hypotheses: first, that models of individual treatment effects would reveal that benefit from intensive blood pressure therapy is proportional to baseline risk; and second, that a machine learning approach designed to predict heterogeneous treatment effects-the X-learner meta-algorithm-is equivalent to a conventional logistic regression approach.
METHODS AND RESULTS: We compared conventional logistic regression to the X-learner approach for prediction of 3-year cardiovascular disease event risk reduction from intensive (target systolic blood pressure <120 mm Hg) versus standard (target <140 mm Hg) blood pressure treatment, using individual participant data from the SPRINT (Systolic Blood Pressure Intervention Trial; N=9361) and ACCORD BP (Action to Control Cardiovascular Risk in Diabetes Blood Pressure; N=4733) trials. Each model incorporated 17 covariates, an indicator for treatment arm, and interaction terms between covariates and treatment. Logistic regression had lower C statistic for benefit than the X-learner (0.51 [95% CI, 0.49-0.53] versus 0.60 [95% CI, 0.58-0.63], respectively). Following the logistic regression's recommendation for individualized therapy produced restricted mean time until cardiovascular disease event of 1065.47 days (95% CI, 1061.04-1069.35), while following the X-learner's recommendation improved mean time until cardiovascular disease event to 1068.71 days (95% CI, 1065.42-1072.08). Calibration was worse for logistic regression; it over-estimated ARR attributable to intensive treatment (slope between predicted and observed ARR of 0.73 [95% CI, 0.30-1.14] versus 1.06 [95% CI, 0.74-1.32] for the X-learner, compared with the ideal of 1). Predicted ARRs using logistic regression were generally proportional to baseline pretreatment cardiovascular risk, whereas the X-learner observed-correctly-that individual treatment effects were often not proportional to baseline risk.
CONCLUSIONS: Predictions for individual treatment effects from trial data reveal that patients may experience ARRs not simply proportional to baseline cardiovascular disease risk. Machine learning methods may improve discrimination and calibration of individualized treatment effect estimates from clinical trial data.
CLINICAL TRIAL REGISTRATION: URL: https://www.clinicaltrials.gov . Unique identifiers: NCT01206062; NCT00000620.

PMID: 30857410 [PubMed - in process]


          Detecting oropharyngeal carcinoma using multispectral, narrow-band imaging and machine learning.      Cache   Translate Page      
Icon for Wiley Related Articles

Detecting oropharyngeal carcinoma using multispectral, narrow-band imaging and machine learning.

Laryngoscope. 2018 11;128(11):2514-2520

Authors: Mascharak S, Baird BJ, Holsinger FC

Abstract
OBJECTIVE: To determine if multispectral narrow-band imaging (mNBI) can be used for automated, quantitative detection of oropharyngeal carcinoma (OPC).
STUDY DESIGN: Prospective cohort study.
METHODS: Multispectral narrow-band imaging and white light endoscopy (WLE) were used to examine the lymphoepithelial tissues of the oropharynx in a preliminary cohort of 30 patients (20 with biopsy-proven OPC, 10 healthy). Low-level image features from five patients were then extracted to train naïve Bayesian classifiers for healthy and malignant tissue.
RESULTS: Tumors were classified by color features with 65.9% accuracy, 66.8% sensitivity, and 64.9% specificity under mNBI. In contrast, tumors were classified with 52.3% accuracy (P = 0.0108), 44.8% sensitivity (P = 0.0793), and 59.9% specificity (P = 0.312) under WLE. Receiver operating characteristic analysis yielded areas under the curve (AUC) of 72.3% and 54.6% for classification under mNBI and WLE, respectively (P = 0.00168). For classification by both color and texture features, AUC under mNBI increased (80.1%, P = 0.00230) but did not improve under WLE (below 55% for both models, P = 0.180). Cross-validation with five folds yielded an AUC above 80% for both mNBI models and below 55% for both WLE models (P = 0.0000410 and 0.000116).
CONCLUSION: Compared to WLE, mNBI significantly enhanced the performance of a naïve Bayesian classifier trained on low-level image features of oropharyngeal mucosa. These findings suggest that automated clinical detection of OPC might be used to enhance surgical vision, improve early diagnosis, and allow for high-throughput screening.
LEVEL OF EVIDENCE: NA. Laryngoscope, 2514-2520, 2018.

PMID: 29577322 [PubMed - indexed for MEDLINE]


          Data Scientist Lead - Schneider National - Green Bay, WI      Cache   Translate Page      
Experience with machine learning software (e.g., R, Python, SPSS, SAS), data access/manipulation (e.g., SQL, pandas, dplyr) and NoSQL databases (e.g., MongoDB,...
From Schneider National - Thu, 03 Jan 2019 06:22:26 GMT - View all Green Bay, WI jobs
          Data Scientist Lead - Schneider - Green Bay, WI      Cache   Translate Page      
Experience with machine learning software (e.g., R, Python, SPSS, SAS), data access/manipulation (e.g., SQL, pandas, dplyr) and NoSQL databases (e.g., MongoDB,...
From Schneider - Wed, 02 Jan 2019 23:36:22 GMT - View all Green Bay, WI jobs
          (USA-MD-Laurel) Senior Software Engineer / Computer Scientist      Cache   Translate Page      
## Position Description **Are you looking for an opportunity that will keep you engaged, challenged, and growing year after year?** **Are you searching for meaningful work that prioritizes creative problem-solving over profits?** **Do you have a strong software engineering and mathematics background?** If so, we're looking for someone like you to join our team at APL. The Tactical Intelligence Systems Group of the Asymmetric Operations Sector is seeking experienced engineers, scientists, and developers driven by curiosity, motivated to deliver solutions, and who have a real passion for learning! We are looking for developers create powerful, cutting-edge solutions for challenges in immersive user interfaces, run-time simulation, machine learning, and artificial intelligence. This may involve: * Surveying academic research to and industry tools to solve problems related to game engine rendering, graphics optimization, and custom shaders * Crafting simulations to generate datasets for machine learning algorithms * Performing full-stack architecture and API design for integrating diverse systems * Building and implementing artificial intelligence algorithms to drive characters in a variety of simulations * Developing immersive user experiences in augmented and virtual reality * Developing software frameworks to manage and analyze agent behavior * Collaborating with Laboratory, for-profit contractor, and sponsor teams to address critical sponsor needs * Effectively communicating results with non-expert audiences, and generating creative ideas to benefit the country. * Some limited travel (up to 10%) to customer sites, and occasional weekend and other after-hours work required to handle and/or complete critical project/work-related business needs. **As a Senior Software Engineer / Computer Scientist, you will....** * Primarily be responsible for applying knowledge in game design, machine learning, full-stack design, and software development to data analysis problems for our sponsors. * Work independently and on teams to engineer software solutions. * Explore promising research and maintain / gain the technical edge required for projects, and share and develop new approaches and methods. * Collaborate to document and support software analytics, and clearly present status and results to internal and external partners. ## Qualifications **You meet our minimum qualifications for the job if you...** * Possess a BS degree in Computer Science, Mathematics, or a related technical track. * Have 5+ years of programming experience and a strong math background. * Are willing and able to deliver operational solutions within business constraints. * Have demonstrated experience in at least three of the following areas: software development, development using 3D game engine, back-end development, machine learning, natural language processing and translation, knowledge representation and reasoning with evidence, synthetic data generation. * Hold an active [Secret or Top Secret] security clearance. If selected, you will be subject to a government security clearance investigation and must meet the requirements for access to classified information. Eligibility requirements include U.S. citizenship. * Are a U.S. Citizen with the ability to obtain a Department of Defense security clearance. If selected, you will be subject to a government security clearance investigation and must meet the requirements for access to classified information. Eligibility requirements include U.S. citizenship. **You'll go above and beyond our minimum requirements if you...** * Have a Master's or PhD degree in Computer Science or a related field and 10+ years of relevant experience. * Possess 2+years of experience performing development using a graphics engine such as Unity3D, Unreal Engine, Blender or Maya. * Have 2+ years of experience applying machine learning to data science or artificial intelligence. * Are experienced in developing Augmented / Virtual Reality solutions. * Are experienced with team-based development of software products and are able to lead development and research projects. * Have experience with machine learning libraries such as Tensorflow, Keras, Caffe, MXNet, CNTK, and scikit-learn, and are familiar with modern databases and parallel computation. * Possess a current DoD security clearance. **Why work at APL?** The Johns Hopkins University Applied Physics Laboratory (APL) brings world-class expertise to our nation’s most critical defense, security, space and science challenges. With a wide selection of challenging, impactful work and a robust education assistance program, APL promotes a culture of life-long learning. Our employees enjoy generous benefits and healthy work/life balance. APL’s campus is located in the Baltimore-Washington metro area. Learn more about our career opportunities at www.jhuapl.edu/careers. APL is an Equal Opportunity/Affirmative Action employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, gender identity, sexual orientation, national origin, disability status, veteran status, or any other characteristic protected by applicable law. *Primary Location:* *United States-*Maryland-*Laurel *Schedule:* Full-time *Req ID:* 18517
          Apple rachète (encore) une société de Machine Learning      Cache   Translate Page      
none
          Artificial Intelligence Developer      Cache   Translate Page      
TX-Round Rock, Artificial Intelligence/ Machine Learning Developer Work Location : Round Rock, TX Start Date : ASAP Duration: 12 Month+ CTH Potential to convert to FTE AI/ML Developer As AI/ML Developer, you work closely with development teams to ensure accurate integration of artificial intelligence/machine learning models into firm platforms. You will contribute to application testing and provide customer focu
          Review: Sophos Intercept X Stops Threats at the Gate      Cache   Translate Page      
Review: Sophos Intercept X Stops Threats at the Gate eli.zimmerman_9856 Tue, 03/12/2019 - 11:59

Traditional anti-malware products scan both memory and disk for particular threat signatures, which are updated daily (or even more often). But if a new threat appears before the pattern files are updated, these solutions won’t be able to detect or prevent the attack. 

In an effort to keep ahead of hackers, SophosLabs analyzes more than 400,000 new malware samples every day. The challenge is that the vast majority of malware is unique to individual organizations, so updating a pattern file is an inefficient, ineffective block for these attacks.

To fix that, Sophos Intercept X sits on top of traditional security software solutions to augment protection. The software prevents malware before it can be executed and stops threats, such as ransomware, from running. When ransomware does get into the network, the tool provides a root cause analysis to help users understand the forensic details.

MORE FROM EDTECH: Here are four ways universities can improve their endpoint protection.

Defeat Ransomware with Automatic Monitoring and File Rollbacks

Intercept X uses deep learning to detect new (and previously unseen) malware and unwanted applications. Deep learning is modeled after the human brain, using advanced neural networks that continuously learn as they accumulate more data.

It’s the same kind of machine learning that powers facial recognition, natural language processing and even self-driving cars, all inside an anti-malware program.

Sheen

Ransomware has grown at a fast clip since the success of the WannaCry malware infection in May 2016. Ransomware installs itself on a computer and then encrypts important files, making them inaccessible to their owner. The owner then receives a message from the attackers that, in an exchange for currency, they will decrypt the files

Sophos Intercept X blocks these attacks by monitoring the file system, detecting any rapid encryption of files and terminating the process. It even rolls back the changes to the files, leaving them as if they had never been touched — and denying the cybercriminals a payoff.

Integrated Protections Give Admins Better Visibility

The software offers several additional protections. WipeGuard uses the same deep learning features to protect a computer’s Master Boot Record. (Ransomware attacks on the MBR prevent the computer from restarting — even restores from backups are impossible until the cybercriminals get their money.)

Safe Browsing includes policies to monitor a web browser’s encryption, presentation and network interfaces to detect “man in the browser” attacks that are common in many banking Trojan viruses.

Sophos Root Cause Analysis contains a list of infection types that have occurred in the past 90 days. There’s even a Visualize tab that connects devices, browsers and websites to track where the infection occurred and how it spread. 

This doesn’t mean users must take action immediately, but it could help them investigate the chain of events surrounding a malware infection and highlight any necessary security improvements.

One caveat: If users haven’t patched their software (especially Java and Adobe applications), Intercept X may detect false positives. Be sure to update all software to the most current versions — always a best practice — to avoid these accidental alerts.

Cybersecurity-report_EasyTarget.jpg

Make Management Easier Through Sophos Central Dashboard

Endpoint protection is wonderful, but managing all those endpoints can be a chore. In addition to the usual laptops and desktops, security managers must pay attention to servers, mobile devices, email and web browsing. The potential threat surface can be overwhelming.

Sophos Central streamlines endpoint management, especially when deployed alongside other Sophos products. From the console, admins can manage Intercept X and endpoint protection either globally or by device. Web protection provides enterprise-grade browsing defense against malicious pop-ups, ads and risky file downloads. The mobile dashboard also shows device compliance, self-service portal registrations, platform versions and management status. 

Server security protects both virtual and physical servers. The Server Lockdown feature reduces the possibility of attack by ensuring that a server can configure and run only known, trusted executables.

Sophos wireless, encryption and email products also tie in to the console, and Sophos Wi-Fi access points can work alongside endpoint and mobile protection clients to provide integrated threat protection. 

That lets admins see what’s happening on wireless networks, APs and connecting clients to get insight into the inappropriate use of resources, including rogue APs. 

The Sophos Encryption dashboard provides centrally managed full-disk encryption using Windows BitLocker or Mac FileVault. Key management becomes a snap with the SafeGuard Management Center, which lets users recover damaged systems. 

Sophos email protection provides a safeguard against spam, phishing attempts and other malicious attacks through the most common user interface of all: email.

Sophos Central isn’t just for admins. Self-service is an important feature today, with user demands and IT budgets in constant conflict. 

Users can log in to the Sophos self-service portal to customize their security status, recover passwords and get notifications. In most IT departments, password recovery is the No. 1 help desk request, and eliminating those calls means technicians can spend more time on complex tasks.

Sophos Intercept X

OS: Windows 7, 8, 8.1 and 10, 32-bit and 64-bit; macOSz
Speed: Extracts millions of file features in 20 millisecondsm
Storage Requirement: 20MB on the endpoint
Server Requirement: Sophos Central supported on Windows 2008R2 and above

Dr. Jeffrey Sheen currently works as the supervisor of enterprise architecture services for Grange Mutual Casualty Group of Columbus, Ohio.


          Apple acquires Laserlike, an ML startup that might make Siri smarter      Cache   Translate Page      
Another small machine learning/AI company acquisition for Apple might mean a smarter Siri -- one capable of parsing the web and delivering personal results.
          How AI helps eBay connect buyers and sellers across 1.2 billion listings      Cache   Translate Page      
In a blog post, eBay highlighted ways it's using machine learning to "enhance experiences" and inspire "economic empowerment" throughout its marketplace.
          (Senior) Engineer - Development and integration of Machine Learning products - SAP - Sankt Leon-Rot      Cache   Translate Page      
Requisition ID: 201307 Work Area: Software-Design and Development Location: Walldorf/St. Leon-Rot Expected Travel: 0 - 10% Career Status: Professional...
Gefunden bei SAP - Tue, 05 Mar 2019 12:36:55 GMT - Zeige alle Sankt Leon-Rot Jobs
          El ecosistema de TensorFlow para programadores principiantes y expertos en Machine Learning: cursos, lenguajes y Edge Computing      Cache   Translate Page      

El ecosistema de TensorFlow para programadores principiantes y expertos en Machine Learning: cursos, lenguajes y Edge Computing#source%3Dgooglier%2Ecom#https%3A%2F%2Fgooglier%2Ecom%2Fpage%2F%2F10000

TensorFlow es la apuesta clave de Google para construir el ecosistema del futuro del Machine Learning que pueda ser ejecutado en la nube, en aplicaciones o en dispositivos hardware de todo tipo.

Precisamente, los esfuerzos en su última TensorFlow Dev Summit 2019 han ido enfocados en facilitar y simplificar el uso del framework, incorporando más API tanto para los programadores principiantes como para los más expertos. De este modo, todos podremos aprovecharnos de las nuevas mejoras para crear modelos de aprendizaje más fácilmente para la mayor número de casos de uso y desplegarlos en cualquier dispositivos.

Han impulsado el despliegue de los algoritmos de forma local en dispositivos hardware con la release final de TensorFlow Lite 1.0 sin necesidad de recurrir a la nube u otro sistema centralizado para ser procesados. Un claro ejemplo de que el Edge Computing forma parte de la estrategia clave de Google para dotar a cualquier dispositivo, ya sea IoT o móvil, de todas las ventajas del aprendizaje automático.

En los tres años que han pasado desde su lanzamiento, TensorFlow ha sentando las bases de un ecosistema de Machine Learning end-to-end, ayudando a potenciar la revolución del Deep Learning. Cada vez hay más desarrolladores que hacen uso de algoritmos para implementar nuevas funcionalidades a los usuarios o acelerar tareas hasta ahora tediosas como la clasificación de imágenes, la captura y reconocimiento de documentos o el reconocimiento de voz y la síntesis del lenguaje natural en los asistentes virtuales (Google Assistant o Alexa)

No es extraño que TensorFlow sea el proyecto con mayor número de contribuciones en Github año tras años, con más de 1.800 contribuciones. Acumulando más de 41 millones de descargas en tres años de historia y decenas de ejemplos de uso en distintas plataformas.

Tensorflow

El camino hacia TensorFlow 2.0

TensorFlow ha sentando las bases de un ecosistema de Machine Learning end-to-end, ayudando a potenciar la revolución del Deep Learning

TensorFlow 2.0 Alpha se ha fijado el objetivo de simplificar su uso, ampliando las posibilidades para ser una plataforma de ML más abierta que puede ser utilizada tanto por investigadores que quieran realizar experimentos y estudios, desarrolladores dispuestos a automatizar cualquier clase tarea o empresas que quieran facilitar la experiencia de uso de sus usuarios a través del inteligencia artificial.

Uno de los pilares de TensorFlow 2.0 es la integración más estrecha con Keras como la API de alto nivel para construir y entrenar modelos de Deep Learning. Keras tiene diversas ventajas:

  • Enfocada en el usuario. Keras tiene una interfaz más simple y consistente, adaptada a los casos de uso más comunes. Facilitando un feedback más claro para entender los errores de implementación.
  • Ser más modular y reutilizable. De este modo los modelos de Keras puede componer estructuras más complejas a través de capas y optimizadores sin necesidad de un modelo específico para entrenar.
  • Pensado tanto para principiantes como para expertos. Aprovechando como idea fundamental el background de los diversos tipos de programadores que se están involucrando desde el principio en el desarrollo de Deep Learning. Keras provee una API mucho más clara sin necesidad de ser un experto con años de experiencia.

También se han incorporado una amplia colección de datasets públicos preparados para ser utilizados con TensorFlow. Cualquier desarrollador que se haya aventurado a trabajar en Machine Learning sabe, esto representa el ingrediente principal para crear modelos y entrenar los algoritmos que después emplearemos. Tener esa ingente cantidad de datos ayuda bastante.

Para la migración de TensorFlow 1.x a 2.0 se han creado diversas herramientas para convertir y migrar los modelos. Ya que se han realizado actualizaciones necesarias para que sean más óptimos y pueden ser desplegados en más plataformas.

El ecosistema sigue creciendo con numerosas librerías para crear un entorno de trabajo más seguro y privado. Como el lanzamiento de la librería TensorFlow Federated, cuya intención es descentralizar el proceso de entrenamiento para compartirlo con múltiples participantes que puedan ejecutarlo localmente y envíen los resultados sin exponer necesariamente los datos capturados, sólo compartiendo el aprendizaje obtenido para la generación de los algoritmos. Un claro ejemplo de esto es el aprendizaje automático de los teclados virtuales, como el de GBoard de Google, que no expone datos sensibles, ya que va a aprendiendo localmente en el propio dispositivo.

Federated Tensor Flow Gboard

Al hilo de esto, el equilibrio entre Machine Learning y la privacidad es una tarea compleja, por ello se ha lanzado la librería de TensorFlow Privacy que permite definir distintos escenarios y grados para salvaguardar los datos más sensible y anonimizar la información de entrenamiento de los modelos.

Python no está solo en TensorFlow, más lenguajes como Swift o Javascript se unen a la plataforma

Python sigue siendo una pieza fundamental en el ecosistema Machine Learning y a la vez ha recibido un gran impulso al ser uno de los lenguajes principales

Obviamente, Python sigue siendo una pieza fundamental en el ecosistema Machine Learning y a la vez ha recibido un gran impulso al ser uno de los lenguajes principales con decenas de librerías entre las más utilizadas, a parte de su gran madurez. No sólo en TensorFlow, sino en otras plataformas como PyTorch.

Pero el ecosistema de TensorFlow ha abierto sus puertas incorporando librerías como TensorFlow.js, que finalmente alcanza la versión 1.0 Con más de 300.000 descargas y 100 contribuciones. Permite ejecutar proyectos ML en el navegador o el en backend con Node.js, tanto modelos ya pre entrenados como construir entrenamientos.

Empresas como Uber o Airbnb ya lo están utilizando en entornos de producción. Hay una amplia galería de ejemplos y casos de uso utilizando JavaScript junto a TensorFlow

Tensor Flow Switf

Otra de las grandes novedades es el avance en la implementación de TensorFlow en Swift con su versión 0.2. De esta forma incorporan un nuevo lenguaje de propósito general como Swift al paradigma ML con todas las funcionalidades para que los desarrolladores puedan acceder a todos los operadores de TensorFlow. Todo ello construido sobre las bases de Jupyter y LLDB.

Ejecutar localmente nuestros modelos con Tensor Lite: la apuesta por el Edge computing

Edge computing se basa en ejecutar modelos y hacer inferencias directamente en local sin depender de que tengan que ser enviados los datos para ser analizados a la nube.

El objetivo de TensorFlow Lite es impulsar definitivamente el Edge Computing en los millones de dispositivos que a día de hoy son capaces de ejecutar TensorFlow. Es la solución para ejecutar modelos y hacer inferencias directamente sin depender de que tengan que ser enviados los datos para ser analizados a la nube.

En Mayo de 2017 fue presentado en la conferencia de desarrolladores de Google IO con una versión preview. Esta semana ha alcanzado la versión definitiva TensorFlow Lite 1.0. Lo que ayudará a implementar diversos casos de uso como la generación de texto predictivo, la clasificación de imágenes, la detección de objetos, el reconocimiento de audio o la síntesis de voz, entre otros muchos escenarios que se pueden implementar.

Esto permite una mejora de rendimiento considerable debido a la conversión al modelo de TensorFlow Lite que permite la herramienta, como por el aumento de rendimiento para ser ejecutado en las GPU de cada dispositivo, incluyendo Android, por ejemplo.

Con ello, TensorFlow Mobile comienza a ser deprecado, salvo que realmente queramos realizar entrenamientos directamente desde el mismo dispositivo. Ya han confirmado que dentro del roadmap de esta versión Lite están trabajando en esto mismo, desvelaron ciertas funcionalidades interesantes como el aprendizaje acelerado con la asignación de pesos para mejorar la inferencia e incorporar ese aprendizaje en sucesivas ejecuciones.

Para completar los novedades, se presentó Google Coral una placa hardware que permite desplegar modelos usando TensorFlow Lite y toda la potencia de la Edge TPU de Google.

Tensor Flow Coral Board

Aprender sobre TensorFlow y Machine Learning cada vez más fácil con estos cursos

En 2016, Udacity lanzó el primer curso sobre TensorFlow en colaboración con Google. Desde entonces, más de 400.000 estudiantes se han apuntado al curso. Aprovechando el lanzamiento de Tensor Flow 2.0 Alpha, se ha renovado el curso por completo para hacerlo más accesible a cualquier desarrollador sin requerir un profundo conocimiento en matemáticas. Tal como ellos afirman: “Si puedes programar, puedes construir aplicaciones AI con Tensor Flow”

Cursos Tensorflow Udacity Deeplearning

El curso de Udacity está guiado por el equipo de desarrollo de Google, a día de hoy está disponible la formación de los primeros 2 meses de la planificación, pero irán añadiendo más contenido a lo largo de las semanas. En la primera parte podrás aprender los conceptos fundamentales detrás del machine learning y cómo construir tu primera red neuronal usando TensorFlow. Disponen de numerosos ejercicios y codelabs escritos por el propio equipo de Tensor Flow.

También se ha incorporado nuevo material en deeplearning.ai con un curso de introducción a AI, ML y DL, parte del career path de Tensor Flow: from Basics to Mastery series de Coursera. Entre los instructores se encuentra Andrew Ng, uno de los más importantes impulsores del Machine Learning desde sus inicios.

Y otra de las plataformas orientadas a la formación en AI, Fast.ai ha incorporado dos cursos sobre el uso de TensorFlow Lite para desarrolladores móviles y otro sobre el uso de Swift en TensorFlow.

Definitivamente, tenemos muchas oportunidades para empezar a aprender más sobre la revolución del Machine Learning junto a TensorFlow, una de las plataformas end-to-end más completas para este fin.

recommendation.header

Gasolina, diésel, híbrido… Cómo acertar con la propulsión

IA y redes neuronales, la apuesta de Google para luchar contra la pornografía infantil

Estados Unidos y Rusia paran las negociaciones sobre un tratado para prohibir los robots asesinos

-
the.news El ecosistema de TensorFlow para programadores principiantes y expertos en Machine Learning: cursos, lenguajes y Edge Computing originally.published.in por Txema Rodríguez .


          This Chrome Extension Lets You Tune Out Toxic Comments Online      Cache   Translate Page      
“Tune” is an app that uses machine learning to filter what it thinks is “toxic” content, but it’s far from perfect.
          Dell EMC, Nvidia make AI reference architecture available      Cache   Translate Page      
AI is becoming a critical workload for enterprises and storage giants are rolling out building blocks to build out machine learning and AI workloads.
          C3 launches Integrated Development Studio for speed up AI, machine learning deployments      Cache   Translate Page      
C3's approach with its development studio is designed to abstract processes across multiple clouds in a low-code environment.
          Um guia essencial para Numpy para Machine Learning em Python      Cache   Translate Page      
Por que isso seria útil para você? Bem, já que a maioria de nós tende a esquecer (no caso daqueles que já implementaram algoritmos ML) as várias funções da biblioteca e acabam escrevendo código para funções pré-existentes usando lógica pura que é um desperdício de tempo e energia, em tais tempos torna-se essencial se entendermos […]
          Grow your games business with ads      Cache   Translate Page      

There’s so much that goes into building a great mobile game. Building a thriving business on top of it? That’s next level. Today, we’re announcing new solutions to increase the lifetime value of your players. Now, it’s easier than ever to re-engage your audience and take advantage of a new, smarter approach to monetization.

Help inactive players rediscover your game

Let's face it, the majority of players you acquire aren't going to continue engaging with your game after just a handful of days. One of the biggest opportunities you have to grow your business is to get those inactive players to come back and play again.

We’re introducing App campaigns for engagement in Google Ads to help players rediscover your game by engaging them with relevant ads across Google’s properties. With App campaigns for engagement, you can reconnect with players in many different ways, such as encouraging lapsed players to complete the tutorial, introducing new features that have been added since a player’s last session, or getting someone to open the game for the first time on Android (which only Google can help with).

Learn more about it here or talk to your Google account representative if you’re interested in trying it out.

Rediscover game 1

Generate revenue from non-spending players

Acquiring and retaining users is important, but retention alone doesn’t generate revenue.  Our internal data shows that, on average, less than four percent of players will ever spend on in-app items. One way to increase overall revenue is through ads. However, some developers worry that ads might hurt in-app purchase revenue by disrupting gameplay for players who do spend. What if you could just show ads to the players who aren't going to spend in your app? Good news—now you can.

We’re bringing a new approach to monetization that combines ads and in-app purchases in one automated solution. Available today, new smart segmentation features in Google AdMob use machine learning to segment your players based on their likelihood to spend on in-app purchases.

Ad units with smart segmentation will show ads only to users who are predicted not to spend on in-app purchases. Players who are predicted to spend will see no ads, and can simply continue playing.  Check it out by creating an interstitial ad unit with smart segmentation enabled.

Smart Segmentation Flow

To learn more about news ways to help you increase the lifetime value of your players, please join us at the Game Developers Conference. Location and details are below:


What: Google Ads Keynote
Where: Moscone West, room #2020
When: Wednesday March 20th at 12:30 PM


I'm excited for the week ahead and all the new games you’re building—I’m always on the lookout for my next favorite.



          Data Engineer (Data Warehouse) - Bandwidth - Raleigh, NC      Cache   Translate Page      
Machine Learning &amp; statistics experience. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Mon, 25 Feb 2019 20:43:17 GMT - View all Raleigh, NC jobs
          Senior Software Engineer, Network Engineering - Bandwidth - Raleigh, NC      Cache   Translate Page      
You've got experience with machine learning and anomaly detection. We need developers who learn quickly and are comfortable diving into the operational...
From Bandwidth - Tue, 19 Feb 2019 16:35:34 GMT - View all Raleigh, NC jobs
          Database Admin Intern - Bandwidth - Raleigh, NC      Cache   Translate Page      
Familiarity with Machine Learning &amp; Statistical concepts. Provide operational support to satisfy internal reporting/data requests....
From Bandwidth - Tue, 20 Nov 2018 22:34:57 GMT - View all Raleigh, NC jobs
          Applied Scientist - Machine Learning - Amazon.com Services, Inc. - Bellevue, WA      Cache   Translate Page      
The Last Mile Science &amp; Technology organization plays a crucial role in developing solutions to vehicle routing problems critical for Amazon Logistics, Prime...
From Amazon.com - Tue, 26 Feb 2019 07:51:35 GMT - View all Bellevue, WA jobs
          Product Manager - CRM Web Solutions, LLC - Bedford, TX      Cache   Translate Page      
Communicate feature releases to internal and external users. Experience in working on AI, Machine Learning products is a huge plus....
From CRM Web Solutions, LLC - Tue, 12 Mar 2019 15:31:15 GMT - View all Bedford, TX jobs
          Machine Learning/AI Engineer - Groom & Associes - Montréal, QC      Cache   Translate Page      
Knowledge of NEON/SSE and vector processing. An international company is looking for Machine Learning/AI Engineers to join their new Montreal office!...
From Groom & Associes - Fri, 08 Mar 2019 21:29:13 GMT - View all Montréal, QC jobs
          Want to manage your total cloud costs better? Emphasize the ‘Ops’ in DevOps, says Futurum analyst Daniel Newman      Cache   Translate Page      

The next BriefingsDirect Voice of the Analyst interview explores new ways that businesses can gain the most control and economic payback from various cloud computing models.

We’ll now hear from an IT industry analyst on how developers and IT operators can find newfound common ground to make hybrid cloud the best long-term economic value for their organizations.

Listen to the podcast. Find it on iTunes. Read a full transcript or download a copy.

Here to help explore ways a managed and orchestrated cloud lifecycle culture should be sought across enterprise IT organizations is Daniel Newman, Principal Analyst and Founding Partner at Futurum Research. The interview is moderated by Dana Gardner, Principal Analyst at Interarbor Solutions.

Here are some excerpts:

Gardner: Daniel, many tools have been delivered over the years for improving software development in the cloud. Recently, containerization and management of containers has been a big part of that.

Now, we’re also seeing IT operators tasked with making the most of cloud, hybrid cloud, and multi-cloud around DevOps – and they need better tools, too.

Has there been a divide or lag between what developers have been able to do in the public cloud environment and what operators must be able to do? If so, is that gap growing or shrinking now that new types of tools for automation, orchestration, and composability of infrastructure and cloud services are arriving?

Out of the shadow, into the cloud 

Newman: Your question lends itself to the concept of shadow IT. The users of this shadow IT find a way to get what they need to get things done. They have had a period of uncanny freedom.

Newman
But this has led to a couple of things. First of all, generally nobody knows what anybody else is doing within the organization. The developers have been able to creatively find tools.

On the other hand, IT has been cast inside of a box. And they say, “Here is the toolset you get. Here are your limitations. Here is how we want you to go about things. These are the policies.”

And in the data center world, that’s how everything gets built. This is the confined set of restrictions that makes a data center a data center.

But in a developer’s world, it’s always been about minimum viable product. It’s been about how to develop using tools that do what they need them to do and getting the code out as quickly as possible. And when it’s all in the cloud, the end-user of the application doesn’t know which cloud it’s running on, they just know they’re getting access to the app.

Basically we now have two worlds colliding. You have a world of strict, confined policies -- and that’s the “ops” side of DevOps. You also have the developers who have been given free rein to do what they need to do; to get what they need to get done, done.

Get Dev and Ops to collaborate 

Gardner: So, we need to keep that creativity and innovation going for the developers so they can satisfy their requirements. At the same time, we need to put in guard rails, to make it all sustainable.

Otherwise we see not a minimal viable cloud – but out-of-control expenses, out-of-control governance and security, and difficulty taking advantage of both private cloud and public cloud, or a hybrid affair, when you want to make that choice.

How do we begin to make this a case of worlds collaborating instead of worlds colliding?

Newman: It’s a great question. We have tended to point DevOps toward “dev.” It’s really been about the development, and the “ops” side is secondary. It’s like capital D, lowercase o.

The thing is, we’re now having a massive shift that requires more orchestration and coordination between these groups.
How to Make
Hybrid IT
Simple
You mentioned out-of-control expenses. I spoke earlier about DevOps and developers having the free rein – to do what they need to do, put it where they need to put it, containers, clouds, tools, whatever they need, and just get it out because that’s what impacts their customers.

If you have an application where people buy things on the web and you need to get that app out, it may be a little more expensive to deploy it without the support of Ops, but you feel the pressure to get it done quickly.

Now, Ops can come in and say, “Well, you know … what about a flex consumption-based model, what about multi-cloud, what about using containers to create more portability?”

“What if we can keep it within the constraints of a budget and work together with you? And, by the way, we can help you understand which applications are running on which cloud and provide you the optimal [aggregate cloud use] plan.”

Let’s be very honest, a developer doesn’t care about all of that. ... They are typically not paid or compensated in any way that leads to optimizing on cost. That’s what the Ops people do.

Such orchestration -- just like almost all larger digital transformation efforts -- starts when you have shared goals. The problem is, they call it a DevOps group -- but Dev has one set of goals and Ops has different ones.

What you’re seeing is the need for new composable tools for cloud services, which we saw at such events as the recent Hewlett Packard Enterprise (HPE) Discover conference. They are launching these tools, giving the Ops people more control over things, and -- by the way -- giving developers more visibility than has existed in the past.
There is a big opportunity [for better cloud use economics] through better orchestration and collaboration, but it comes down to the age-old challenges of having the Dev and Ops people share the same goals.

There is a big opportunity [for better cloud use economics] through better orchestration and collaboration, but it comes down to the age-old challenges inside of any IT organization -- and that is having the Dev and the Ops people share the same goals. These new tools may give them more of a reason to start working in that way.

Gardner: The more composability the operations people have, the easier it is for them to define a path that the developers can stay inside of without encumbering the developers.

We may be at the point in the maturity of the industry where both sides can get what they want. It’s simply a matter of putting that together -- the chocolate and peanut-butter, if you will. It becomes more of a complete DevOps.

But there is another part of this people often don’t talk about, and that’s the data placement component. When we examine the lifecycle of a modern application, we’re not just developing it and staging it where it stays static. It has to be built upon and improved, we are doing iterations, we are doing Agile methods.

We also have to think about the data the application is consuming and creating in the same way. That dynamic data use pattern needs to fit into a larger data management philosophy and architecture that includes multi-cloud support.

I think it’s becoming DevDataOps-- not just DevOps these days. The operations people need to be able to put in requirements about how that data is managed within the confines of that application’s deployment, yet kept secure, and in compliance with regulations and localization requirements.

DevDataOps emerges

Newman: We’ve launched the DevDataOps category right now! That’s actually a really great point, because if you think about where does all that live -- meaning IT orchestration of the infrastructure choices and whether that’s in the cloud or on-premises – there has to be enough of the right kind of storage.

Developers are usually worried about data from the sense of what can they do with that data to improve and enhance the applications. When you add in elements like machine learning (ML) and artificial intelligence (AI), that’s going to just up the compute and storage requirements. You have the edge and Internet of Things (IoT) to consider now too for data. Most applications are collecting more data in real-time. With all of these complexities, you have to ask, “Who really owns this data?”

Well, the IT part of DevOps, the “Ops,” typically worries about capacity and resources performance for data. But are they really worried about the data in these new models? It brings in that needed third category because the Dev person doesn’t necessarily deal with the data lifecycle. The need to best use that data is a business unit imperative, a marketing-level issue, a sales-level data requirement. It can include all the data that’s created inside of a cloud instance of SAP or Salesforce.
How to Solve Cost
and Utilization Challenges
of Hybrid Cloud
Just think about how many people need to be involved in orchestration to maximize that? Culturally speaking, it goes back to shared tools, shared visibility, and shared goals. It’s also now about more orchestration required across more external groups. So your DevOps group just got bigger, because the data deluge is going to be the most valuable resource any company has. It will be, if it isn’t already today, the most influential variable in what your company becomes.

You can’t just leave that to developers and operators of IT. It becomes core to business unit leadership, and they need to have an impact. The business leadership should be asking, “We have all this data. What are we doing with it? How are we managing it? Where does it live? How do we pour it between different clouds? What stays on-premises and what goes off? How do we govern it? How can we have governance over privacy and compliance?”

I would say most companies really struggle to keep up with compliance because there are so many rules about what kind of data you have, where it can live, how it should be managed, and how long it should be stored.


I think you bring up a great point, Dana. I could probably rattle on about this for a long, long time. You’ve just added a whole new element to DevOps, right here on this podcast. I don’t know that it has to do with specifically Dev or Ops, but I think it’s Dev+Ops+Data -- a new leadership element for meaningful digital transformation.

Gardner: We talked about trying to bridge the gap between development and Ops, but I think there are other gaps, too. One is between data lifecycle management – for backup and recovery and making it the lowest cost storage environment, for example. Then there is the other group of data scientists who are warehousing that data, caching it, and grabbing more data from outside, third-party sources to do more analytics for the entire company. But these data strategies are too often still divorced.

These data science people and what the developers and operators are doing aren’t necessarily in sync. So, we might have another category, which would be Dev+Data+DataScience+Ops.

Add Data Analytics to the Composition 

Newman: Now we’re going four groups. You are firstly talking about the data from the running applications. That’s managed through pure orchestration in DevOps, and that works fine through composability tools. Those tools provide IT the capability to add guard rails to the developers, so they are not doing things in the shadows, but instead do things in coordination.

The other data category is that bigger analytical data. It includes open data, third-party data, and historical data that’s been collected and stored inside of instances of Enterprise resource planning (ERP) apps and Customer-relationship management (CRM) apps for 20 or 30 years. It’s a gold mine of information. Now we have to figure out an extract process and incorporate that data into almost every enterprise-level application that developers are building. Right now Dev and Ops don’t really have a clue what is out there and available across that category because that’s being managed somewhere else, through an analytics group of the company.

Gardner: Or, developers will have to create an entirely different class of applications for analytics alone, as well as integrating the analytics services into all of the existing apps.

Newman: One of the HPE partners I’ve worked with the in the past, SAS, and companies such as SAS and SAP, are going to become much closer aligned with infrastructure. Your DevOps is going to become your analytics Ops, too.
How to Achieve
Composability
Across Your Data Center
Hardware companies have built software apps to run their hardware, but they haven’t been historically building software apps to run the data that sits on the hardware. That’s been managed by the businesses running business intelligence software, such as the ones I mentioned.

There is an opportunity for a new level of coordination to take place at the vendor level, because when you see these alliances, and you see these partnerships, this isn’t new. But, seeing it done in a way that’s about getting the maximum amount of usable data from one system into every application -- that’s futuristic, and it needs to be worked on today.

Gardner: The bottom line is that there are many moving parts of IT that remain disjointed. But we are at the point now with composability and automation of getting an uber-view over services and processes to start making these new connections – technically, culturally, and organizationally.

What I have seen from HPE around the HPE Composable Cloud vision moves a big step in that direction. It might be geared toward operators, but, ultimately it’s geared toward the entire enterprise, and gives the business an ability to coordinate, manage, and gain insights into all these different facets of a digital business.
Companies right now still struggle with the resources to run multi-cloud. They tend to have maybe one public cloud and their on-premises operations. They don't know which is the best cloud approach because they are not getting the total information.

Newman: We’ve been talking about where things can go, and it’s exciting. But let’s take a step back.

Multi-cloud is a really great concept. Hyper-converged infrastructure, it’s all really nice, and there has been massive movement in this area in the last couple of years. Companies right now still struggle with the resources to run multi-cloud. They tend to have maybe one public cloud and their on-premise operations. They have their own expertise, and they have endless contracts and partnerships.

They don’t know which the best-cloud approach is because they are not necessarily getting that total information. It depends on all of the relationships, the disparate resources they have across Dev and Ops, and the data can change on a week-to-week basis. One cloud may have been perfect a month ago, yet all of a sudden you change the way an application is running and consuming data, and it’s now in a different cloud.


What HPE is doing with HPE Composable Cloud takes the cloud plus composable infrastructure and, working through HPE OneSphere and HPE OneView, brings them all into a single view. We’re in a software and user experience world.

The tools that deliver the most usable and valuable dashboard-type of cloud use data in one spot are going to win the battle. You need that view in front of you for quick deployment, with quick builds, portability, and container management. HPE is setting itself in a good position for how we do this in one place.
How to Remove
Complexity From
Multi-Cloud and Hybrid IT
Give me one view, give me my one screen to look at, and I think your Dev and Ops -- and everybody in between – and all your new data and data science friends will all appreciate that view. HPE is on a good track, and I look forward to seeing what they do in the future.


          Support Tools Developer - Pure Storage - Lehi, UT      Cache   Translate Page      
The world is experiencing a revolution driven by next-generation technology like AI, machine learning, virtual reality, quantum computing, and self-driving cars...
From Pure Storage - Thu, 28 Feb 2019 00:41:54 GMT - View all Lehi, UT jobs
          Knowledge Manager - Pure Storage - Lehi, UT      Cache   Translate Page      
The world is experiencing a revolution driven by next-generation technology like AI, machine learning, virtual reality, quantum computing, and self-driving cars...
From Pure Storage - Fri, 22 Feb 2019 22:51:05 GMT - View all Lehi, UT jobs
          Technical Support Engineer I - 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, 23 Dec 2018 08:19:51 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 - Fri, 30 Nov 2018 08:21:49 GMT - View all Lehi, UT jobs
          Senior Data Scientist - Predictive Enterprise Group - Neudesic LLC - Philadelphia, PA      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 - Sat, 15 Dec 2018 21:58:12 GMT - View all Philadelphia, PA 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
          NSBE Intern- National Society of Black Engineers 2019 National Convention - Visa - Detroit, MI      Cache   Translate Page      
Strong research experiences and publication record in machine learning and/or data mining, advanced cryptography, systems security, blockchain and quantum...
From Visa - Wed, 09 Jan 2019 04:11:40 GMT - View all Detroit, MI jobs
          Sr. Security Analyst II - AbbVie - Lake County, IL      Cache   Translate Page      
Understanding of Machine Learning. Coordinate efforts among multiple business units during Response. Interpret and summarize technical information for...
From AbbVie - Thu, 10 Jan 2019 21:07:17 GMT - View all Lake County, IL jobs
          Manager, Sales Development - Pure Storage - Mountain View, CA      Cache   Translate Page      
The world is experiencing a revolution driven by next-generation technology like AI, machine learning, virtual reality, quantum computing, and self-driving cars...
From Pure Storage - Wed, 13 Mar 2019 18:43:33 GMT - View all Mountain View, CA jobs
          An early look at Facebook's Codec Avatars, extremely life-like representations of individuals made by applying machine learning to data collected in a studio (Peter Rubin/Wired)      Cache   Translate Page      

Peter Rubin / Wired:
An early look at Facebook's Codec Avatars, extremely life-like representations of individuals made by applying machine learning to data collected in a studio  —  “There's this big, ugly sucker at the door,” the young woman says, her eyes twinkling, “and he said, ‘Who do you think you are, Lena Horne?’


          NOC Service Desk Analyst II - Apptio - Morrisville, NC      Cache   Translate Page      
Skype for Business, WebEx). Datacenter and server hardware (Red Hat Linux). Apptio's software uses machine learning to translate technology costs and...
From Apptio - Fri, 22 Feb 2019 02:32:52 GMT - View all Morrisville, NC jobs
          04-05-2019 Mahsa Shoaran Seminar - Friday, April 5th @ 2PM in EEB 248       Cache   Translate Page      
Speaker: Mahsa Shoaran, Cornell University Talk Title: Ultra-Low-Power Neural Interfaces: from Monitoring to Diagnosis and Therapy Abstract: Implantable and wearable medical devices are increasingly being developed as alternative therapies for intractable diseases. In particular, undertreated neurological disorders such as epilepsy, migraine, and Alzheimer's disease are of major public health concern around the world, driving the need to explore such new approaches. Despite significant advances in neural interface systems, the small number of recording channels in existing technology remains a barrier to their therapeutic potential. This is mainly due to the fact that simultaneous recording from a large number of electrodes imposes stringent energy and area constraints on the integrated circuits that interface with these electrodes. In this talk, I will first discuss an efficient compressive sensing framework for multichannel cortical implants. Next, I will present the design of our sub-microwatt per channel closed-loop seizure control device and both its in-vivo and offline performance. I will then discuss our latest work on the integration of machine learning algorithms for on-chip classification of neural data. Finally, I will give examples of how these results may be used towards designing new devices, to enhance the lives of millions of people suffering from disabling neurological conditions in future. Biography: Mahsa Shoaran is currently an Assistant Professor in the School of Electrical and Computer Engineering at Cornell University. Prior to joining Cornell, she was a postdoctoral fellow in Electrical Engineering and Medical Engineering at the California Institute of Technology. She received her PhD from EPFL in 2015 and her B.Sc. and M.Sc. from Sharif University of Technology. Her research interests broadly include circuit, system, and algorithm design for diagnostic and therapeutic applications. Mahsa is a recipient of the 2019 Google Faculty Research Award, the Early and Advanced Swiss National Science Foundation Postdoctoral Fellowships, and the NSF Award for Young Professionals Contributing to Smart and Connected Health. She was named a Rising Star in EECS by MIT in 2015. Host: ECE-Electrophysics
          Any better than Clever Hans? Putting AI systems to the test      Cache   Translate Page      

As Machine Learning advances, how will we know that these systems are making the right decisions? In this story from the Fraunhofer Heinrich Hertz Institute, researchers are looking at the learning strategies of today's Ai systems. "We were very surprised by just how broad the range of learned problem-solving strategies is. Even modern AI systems have not always found a meaningful approach, at least not from a human perspective, instead sometimes adopting what we call ‘Clever Hans’ strategies."

The post Any better than Clever Hans? Putting AI systems to the test appeared first on insideHPC.


          Regional Sales Manager - Black & Veatch Family of Companies - United States      Cache   Translate Page      
Business Unit Sector :. Experience with AI, predictive analytics and machine learning landscape is a huge plus....
From Atonix Digital - Wed, 06 Mar 2019 10:38:44 GMT - View all United States jobs
          Introducing Android Q Beta      Cache   Translate Page      

Posted by Dave Burke, VP of Engineering

In 2019, mobile innovation is stronger than ever, with new technologies from 5G to edge to edge displays and even foldable screens. Android is right at the center of this innovation cycle, and thanks to the broad ecosystem of partners across billions of devices, Android's helping push the boundaries of hardware and software bringing new experiences and capabilities to users.

As the mobile ecosystem evolves, Android is focused on helping users take advantage of the latest innovations, while making sure users' security and privacy are always a top priority. Building on top of efforts like Google Play Protect and runtime permissions, Android Q brings a number of additional privacy and security features for users, as well as enhancements for foldables, new APIs for connectivity, new media codecs and camera capabilities, NNAPI extensions, Vulkan 1.1 support, faster app startup, and more.

Today we're releasing Beta 1 of Android Q for early adopters and a preview SDK for developers. You can get started with Beta 1 today by enrolling any Pixel device (including the original Pixel and Pixel XL, which we've extended support for by popular demand!) Please let us know what you think! Read on for a taste of what's in Android Q, and we'll see you at Google I/O in May when we'll have even more to share.

Building on top of privacy protections in Android

Android was designed with security and privacy at the center. As Android has matured, we've added a wide range of features to protect users, like file-based encryption, OS controls requiring apps to request permission before accessing sensitive resources, locking down camera/mic background access, lockdown mode, encrypted backups, Google Play Protect (which scans over 50 billion apps a day to identify potentially harmful apps and remove them), and much more. In Android Q, we've made even more enhancements to protect our users. Many of these enhancements are part of our work in Project Strobe.

Giving users more control over location

With Android Q, the OS helps users have more control over when apps can get location. As in prior versions of the OS, apps can only get location once the app has asked you for permission, and you have granted it.

One thing that's particularly sensitive is apps' access to location while the app is not in use (in the background). Android Q enables users to give apps permission to see their location never, only when the app is in use (running), or all the time (when in the background).

For example, an app asking for a user's location for food delivery makes sense and the user may want to grant it the ability to do that. But since the app may not need location outside of when it's currently in use, the user may not want to grant that access. Android Q now offers this greater level of control. Read the developer guide for details on how to adapt your app for this new control. Look for more user-centric improvements to come in upcoming Betas. At the same time, our goal is to be very sensitive to always give developers as much notice and support as possible with these changes.

More privacy protections in Android Q

Beyond changes to location, we're making further updates to ensure transparency, give users control, and secure personal data.

In Android Q, the OS gives users even more control over apps, controlling access to shared files. Users will be able to control apps' access to the Photos and Videos or the Audio collections via new runtime permissions. For Downloads, apps must use the system file picker, which allows the user to decide which Download files the app can access. For developers, there are changes to how your apps can use shared areas on external storage. Make sure to read the Scoped Storage changes for details.

We've also seen that users (and developers!) get upset when an app unexpectedly jumps into the foreground and takes over focus. To reduce these interruptions, Android Q will prevent apps from launching an Activity while in the background. If your app is in the background and needs to get the user's attention quickly -- such as for incoming calls or alarms -- you can use a high-priority notification and provide a full-screen intent. See the documentation for more information.

We're limiting access to non-resettable device identifiers, including device IMEI, serial number, and similar identifiers. Read the best practices to help you choose the right identifiers for your use case, and see the details here. We're also randomizing the device's MAC address when connected to different Wi-Fi networks by default -- a setting that was optional in Android 9 Pie.

We are bringing these changes to you early, so you can have as much time as possible to prepare. We've also worked hard to provide developers detailed information up front, we recommend reviewing the detailed docs on the privacy changes and getting started with testing right away.

New ways to engage users

In Android Q, we're enabling new ways to bring users into your apps and streamlining the experience as they transition from other apps.

Foldables and innovative new screens

Foldable devices have opened up some innovative experiences and use-cases. To help your apps to take advantage of these and other large-screen devices, we've made a number of improvements in Android Q, including changes to onResume and onPause to support multi-resume and notify your app when it has focus. We've also changed how the resizeableActivity manifest attribute works, to help you manage how your app is displayed on foldable and large screens. To you get started building and testing on these new devices, we've been hard at work updating the Android Emulator to support multiple-display type switching -- more details coming soon!

Sharing shortcuts

When a user wants to share content like a photo with someone in another app, the process should be fast. In Android Q we're making this quicker and easier with Sharing Shortcuts, which let users jump directly into another app to share content. Developers can publish share targets that launch a specific activity in their apps with content attached, and these are shown to users in the share UI. Because they're published in advance, the share UI can load instantly when launched.

The Sharing Shortcuts mechanism is similar to how App Shortcuts works, so we've expanded the ShortcutInfo API to make the integration of both features easier. This new API is also supported in the new ShareTarget AndroidX library. This allows apps to use the new functionality, while allowing pre-Q devices to work using Direct Share. You can find an early sample app with source code here.

Settings Panels

You can now also show key system settings directly in the context of your app, through a new Settings Panel API, which takes advantage of the Slices feature that we introduced in Android 9 Pie.

A settings panel is a floating UI that you invoke from your app to show system settings that users might need, such as internet connectivity, NFC, and audio volume. For example, a browser could display a panel with connectivity settings like Airplane Mode, Wi-Fi (including nearby networks), and Mobile Data. There's no need to leave the app; users can manage settings as needed from the panel. To display a settings panel, just fire an intent with one of the new Settings.Panel actions.

Connectivity

In Android Q, we've extended what your apps can do with Android's connectivity stack and added new connectivity APIs.

Connectivity permissions, privacy, and security

Most of our APIs for scanning networks already require COARSE location permission, but in Android Q, for Bluetooth, Cellular and Wi-Fi, we're increasing the protection around those APIs by requiring the FINE location permission instead. If your app only needs to make peer-to-peer connections or suggest networks, check out the improved Wi-Fi APIs below -- they simplify connections and do not require location permission.

In addition to the randomized MAC addresses that Android Q provides when connected to different Wi-Fi networks, we're adding new Wi-Fi standard support, WP3 and OWE, to improve security for home and work networks as well as open/public networks.

Improved peer-to-peer and internet connectivity

In Android Q we refactored the Wi-Fi stack to improve privacy and performance, but also to improve common use-cases like managing IoT devices and suggesting internet connections -- without requiring the location permission.

The network connection APIs make it easier to manage IoT devices over local Wi-Fi, for peer-to-peer functions like configuring, downloading, or printing. Apps initiate connection requests indirectly by specifying preferred SSIDs & BSSIDs as WiFiNetworkSpecifiers. The platform handles the Wi-Fi scanning itself and displays matching networks in a Wi-Fi Picker. When the user chooses, the platform sets up the connection automatically.

The network suggestion APIs let apps surface preferred Wi-Fi networks to the user for internet connectivity. Apps initiate connections indirectly by providing a ranked list of networks and credentials as WifiNetworkSuggestions. The platform will seamlessly connect based on past performance when in range of those networks.

Wi-Fi performance mode

You can now request adaptive Wi-Fi in Android Q by enabling high performance and low latency modes. These will be of great benefit where low latency is important to the user experience, such as real-time gaming, active voice calls, and similar use-cases.

To use the new performance modes, call WifiManager.WifiLock.createWifiLock() with WIFI_MODE_FULL_LOW_LATENCY or WIFI_MODE_FULL_HIGH_PERF. In these modes, the platform works with the device firmware to meet the requirement with lowest power consumption.

Camera, media, graphics

Dynamic depth format for photos

Many cameras on mobile devices can simulate narrow depth of field by blurring the foreground or background relative to the subject. They capture depth metadata for various points in the image and apply a static blur to the image, after which they discard the depth metadata.

Starting in Android Q, apps can request a Dynamic Depth image which consists of a JPEG, XMP metadata related to depth related elements, and a depth and confidence map embedded in the same file on devices that advertise support.

Requesting a JPEG + Dynamic Depth image makes it possible for you to offer specialized blurs and bokeh options in your app. You can even use the data to create 3D images or support AR photography use-cases in the future. We're making Dynamic Depth an open format for the ecosystem, and we're working with our device-maker partners to make it available across devices running Android Q and later.

With Dynamic Depth image you can offer specialized blurs and bokeh options in your app.

New audio and video codecs

Android Q introduces support for the open source video codec AV1. This allows media providers to stream high quality video content to Android devices using less bandwidth. In addition, Android Q supports audio encoding using Opus - a codec optimized for speech and music streaming, and HDR10+ for high dynamic range video on devices that support it.

The MediaCodecInfo API introduces an easier way to determine the video rendering capabilities of an Android device. For any given codec, you can obtain a list of supported sizes and frame rates using VideoCodecCapabilities.getSupportedPerformancePoints(). This allows you to pick the best quality video content to render on any given device.

Native MIDI API

For apps that perform their audio processing in C++, Android Q introduces a native MIDI API to communicate with MIDI devices through the NDK. This API allows MIDI data to be retrieved inside an audio callback using a non-blocking read, enabling low latency processing of MIDI messages. Give it a try with the sample app and source code here.

ANGLE on Vulkan

To enable more consistency for game and graphics developers, we are working towards a standard, updateable OpenGL driver for all devices built on Vulkan. In Android Q we're adding experimental support for ANGLE on top of Vulkan on Android devices. ANGLE is a graphics abstraction layer designed for high-performance OpenGL compatibility across implementations. Through ANGLE, the many apps and games using OpenGL ES can take advantage of the performance and stability of Vulkan and benefit from a consistent, vendor-independent implementation of ES on Android devices. In Android Q, we're planning to support OpenGL ES 2.0, with ES 3.0 next on our roadmap.

We'll expand the implementation with more OpenGL functionality, bug fixes, and performance optimizations. See the docs for details on the current ANGLE support in Android, how to use it, and our plans moving forward. You can start testing with our initial support by opting-in through developer options in Settings. Give it a try today!

Vulkan everywhere

We're continuing to expand the impact of Vulkan on Android, our implementation of the low-overhead, cross-platform API for high-performance 3D graphics. Our goal is to make Vulkan on Android a broadly supported and consistent developer API for graphics. We're working together with our device manufacturer partners to make Vulkan 1.1 a requirement on all 64-bit devices running Android Q and higher, and a recommendation for all 32-bit devices. Going forward, this will help provide a uniform high-performance graphics API for apps and games to use.

Neural Networks API 1.2

Since introducing the Neural Networks API (NNAPI) in 2017, we've continued to expand the number of operations supported and improve existing functionality. In Android Q, we've added 60 new ops including ARGMAX, ARGMIN, quantized LSTM, alongside a range of performance optimisations. This lays the foundation for accelerating a much greater range of models -- such as those for object detection and image segmentation. We are working with hardware vendors and popular machine learning frameworks such as TensorFlow to optimize and roll out support for NNAPI 1.2.

Strengthening Android's Foundations

ART performance

Android Q introduces several new improvements to the ART runtime which help apps start faster and consume less memory, without requiring any work from developers.

Since Android Nougat, ART has offered Profile Guided Optimization (PGO), which speeds app startup over time by identifying and precompiling frequently executed parts of your code. To help with initial app startup, Google Play is now delivering cloud-based profiles along with APKs. These are anonymized, aggregate ART profiles that let ART pre-compile parts of your app even before it's run, giving a significant jump-start to the overall optimization process. Cloud-based profiles benefit all apps and they're already available to devices running Android P and higher.

We're also continuing to make improvements in ART itself. For example, in Android Q we've optimized the Zygote process by starting your app's process earlier and moving it to a security container, so it's ready to launch immediately. We're storing more information in the app's heap image, such as classes, and using threading to load the image faster. We're also adding Generational Garbage Collection to ART's Concurrent Copying (CC) Garbage Collector. Generational CC is more efficient as it collects young-generation objects separately, incurring much lower cost as compared to full-heap GC, while still reclaiming a good amount of space. This makes garbage collection overall more efficient in terms of time and CPU, reducing jank and helping apps run better on lower-end devices.

Security for apps

BiometricPrompt is our unified authentication framework to support biometrics at a system level. In Android Q we're extending support for passive authentication methods such as face, and adding implicit and explicit authentication flows. In the explicit flow, the user must explicitly confirm the transaction in the TEE during the authentication. The implicit flow is designed for a lighter-weight alternative for transactions with passive authentication. We've also improved the fallback for device credentials when needed.

Android Q adds support for TLS 1.3, a major revision to the TLS standard that includes performance benefits and enhanced security. Our benchmarks indicate that secure connections can be established as much as 40% faster with TLS 1.3 compared to TLS 1.2. TLS 1.3 is enabled by default for all TLS connections. See the docs for details.

Compatibility through public APIs

Another thing we all care about is ensuring that apps run smoothly as the OS changes and evolves. Apps using non-SDK APIs risk crashes for users and emergency rollouts for developers. In Android Q we're continuing our long-term effort begun in Android P to move apps toward only using public APIs. We know that moving your app away from non-SDK APIs will take time, so we're giving you advance notice.

In Android Q we're restricting access to more non-SDK interfaces and asking you to use the public equivalents instead. To help you make the transition and prevent your apps from breaking, we're enabling the restrictions only when your app is targeting Android Q. We'll continue adding public alternative APIs based on your requests; in cases where there is no public API that meets your use case, please let us know.

It's important to test your apps for uses of non-SDK interfaces. We recommend using the StrictMode method detectNonSdkApiUsage() to warn when your app accesses non-SDK APIs via reflection or JNI. Even if the APIs are exempted (grey-listed) at this time, it's best to plan for the future and eliminate their use to reduce compatibility issues. For more details on the restrictions in Android Q, see the developer guide.

Modern Android

We're expanding our efforts to have all apps take full advantage of the security and performance features in the latest version of Android. Later this year, Google Play will require you to set your app's targetSdkVersion to 28 (Android 9 Pie) in new apps and updates. In line with these changes, Android Q will warn users with a dialog when they first run an app that targets a platform earlier than API level 23 (Android Marshmallow). Here's a checklist of resources to help you migrate your app.

We're also moving the ecosystem toward readiness for 64-bit devices. Later this year, Google Play will require 64-bit support in all apps. If your app uses native SDKs or libraries, keep in mind that you'll need to provide 64-bit compliant versions of those SDKs or libraries. See the developer guide for details on how to get ready.

Get started with Android Q Beta

With important privacy features that are likely to affect your apps, we recommend getting started with testing right away. In particular, you'll want to enable and test with Android Q storage changes, new location permission states, restrictions on background app launch, and restrictions on device identifiers. See the privacy documentation for details.

To get started, just install your current app from Google Play onto a device or Android Virtual Device running Android Q Beta and work through the user flows. The app should run and look great, and handle the Android Q behavior changes for all apps properly. If you find issues, we recommend fixing them in the current app, without changing your targeting