ml.ethics

The O'Reilly Data Show explores the opportunities and techniques driving big data and data science. Through interviews and analysis, we highlight the people putting data to work.

The O'Reilly Data Show Podcast

Technology

data

machine learning

Dreaming big is harder than you'd think. 

#2 Is Podcasting the Future or the Past?

Not surprisingly, Joseph Sirosh, has big ambitions for his product portfolio at Microsoft which includes Azure ML, HDInsight and other tools. Chief among them is making it easy for mere mortals to consume these data services from the applications they're familiar with. Take Excel for example.

If a financial analyst can, with a few clicks, send data to a forecast service in the cloud, then get the numbers back, visualized on the same spreadsheet, that's a pretty powerful story, said Sirosh who is corporate VP of machine learning for Microsoft.

But as valuable as those applications and services are, more and more of the value to be derived from computation over time will be the data itself, not all those tech underpinnings.  "In the future a huge part of the value generated from computing will come from the data as opposed to storage and operating systems and basic infrastructure," he noted on this week's podcast. WHich is why one topic under discussion at next month's Structure Data show will be who owns all the data flowing betwixt and betweeen various systems, the internet of things etc.

When it comes to getting corporations running these new systems Microsoft may have an ace in the hole because so many of them already use key Microsoft tools -- Active Directory, SQL Server, Excel. That gives them a pretty good on-ramp to Microsoft Azure and its resident services. Sirosh makes a compelling case and we'll talk to him more on stage at Structure Data next month in New York City.

In the first half of the show, Derrick Harris and I talk about the Hadoop world has returned to its feisty and oh so interesting roots. When Pivotal announced its plan to offload support of Hadoop to Hortonworks and work with that company along with IBM, GE on  the Open Data Platform the response from Cloudera CEO Mike Olsen in a blog post with his take. 

Also on the docket, @WalmartLabs massive OpenStack production private cloud implementation.

Microsoft's machine learning guru on why data matters sooooo much

By John G. Cramer, University of Washington: https://faculty.washington.edu/jcramer/BBSound.html

Related story: http://www.theatlantic.com/science/archive/1857/11/a-brief-history-of-noise/422481/

Frequency Spectrum From the Big Bang (WMAP Satellite Data)

Peter Gasston (https://twitter.com/stopsatgreen) is a front-end developer and web technologist that works for +rehabstudio (https://www.rehabstudio.com) in London.

Read the blog post about this interview: https://blog.feryn.eu/peter-gasston-commoditization-of-artificial-intelligence. The video is up on Youtube: https://youtu.be/_qremcORl9A

This interview was recorded in Rome at Codemotion 2016 (http://www.codemotionworld.com) where Peter presented his "OK Computer" presentation (https://speakerdeck.com/stopsatgreen/ok-computer-codemotion-rome) about the Commoditization of Artificial Intelligence.

We talk about how improved algorithms and the availability of massive computing power due to maturity of Cloud Computing has boosted Artificial Intelligence in a very profound way.

Both major players (Facebook, Google, Microsoft, IBM) and smaller startups can leverage the Cloud for machine learning and offer added value through easy-to-use customer-facing APIs.

Most implementations focus on natural language understanding and computer vision. Here's a list of tools, services and APIs that were mentioned in the interview and/or Peter's presentation.

https://wit.ai
https://cloud.google.com/vision/
http://www.ibm.com/smarterplanet/us/en/ibmwatson/what-is-watson.html
https://cognitoys.com
http://clarif.ai
https://www.metamind.io
http://www.skymind.io
https://www.projectoxford.ai

Thanks to machine learning and massive data processing, artificial intelligence is now a commodity: recognizing objects, locations and people on pictures. Or even understanding the meaning of a conversation. 

► Subscribe to my Youtube channel - http://youtube.com/ThijsFeryn
► Listen to the podcast on iTunes - http://itunes.feryn.eu
► Follow me on Twitter - http://twitter.com/ThijsFeryn
► Read my blog - http://blog.feryn.eu
► Check my presentations - http://talks.feryn.eu
► Thanks to Combell for sponsoring the show - http://www.combell.com

Peter Gasston - Commoditization of Artificial Intelligence

In this episode of the O’Reilly Data Show, I spoke with Mikio Braun, delivery lead and data scientist at Zalando. After spending previous years in academia, Braun recently made the decision to switch to industry. He shared some observations about building large-scale systems, particularly deploying data applications in production systems. Given his longstanding background as a machine learning researcher and practitioner, I wanted to get his take on topics like deep learning, hybrid systems, feature engineering, and AI applications.

Practical Machine Learning Techniques for Building Intelligent Applications

Moore's Law -- putting more and more transistors on a chip -- accelerated the computing industry by so many orders of magnitude, it has (and continues to) achieve seemingly impossible feats. However, we're now resorting to brute-force hacks to keep pushing it beyond its limits and are getting closer to the point of diminishing returns (especially given costly manufacturing infrastructure). Yet this very dynamic is leading to "a Cambrian explosion" in computing capabilities… just look at what's happening today with GPUs, FPGAs, and neuromorphic chips. Through such continuing performance improvements and parallelization, classic computing continues to reshape the modern world.

But we're so focused on making our computers do more that we're not talking enough about what classic computers can't do -- and that's to compute things the way nature does, which operates in quantum mechanics. So our smart machines are really quite dumb, argues Rigetti Computing founder and CEO Chad Rigetti; they're limited to human-made binary code vs. the natural reality of continuous variables. This in turn limits our ability to work on problems that classic computers can't solve, such as key applications in computational chemistry or large-scale optimization for machine learning and artificial intelligence. Which is where quantum computing comes in.

But what is quantum computing, really -- beyond the history and the hype? And where are we in reaching the promise of practical quantum computers? (Hint: it will take a hybrid approach to get there.) Who are the players -- companies, countries, types of people/skills -- working on it, and how can a startup compete in this space? Finally, what will it take to get "the flywheel" of application development and discovery going? Part of the answer comes full circle to the same economic engine that drove previous computing advances, argues Chris Dixon; Moore's Law, after all, is more of an economic principle that combined the forces of capitalism, a critical mass of ideas, and people moving things forward by sheer will. Quantum computing is finally getting pulled into the same economic forces as well.

a16z Podcast: Quantum Computing, Now and Next

In this episode of the Data Show, I spoke with Bruno Fernandez-Ruiz, co-founder and CTO of Nexar. We first met when he was leading Yahoo! technical teams charged with delivering a variety of large-scale, real-time data products. His new company is helping build out critical infrastructure for the emerging transportation sector.

Vehicle-to-Vehicle Communication Networks Can Help Fuel Smart Cities

This week, AI in Industry features Jeremy Barnes, Chief Architect at Element AI. Jeremy talks about the common mistakes some businesses might make while adopting AI to solve broad business problems. He also sheds light on the problem areas that could raise the market value of businesses through AI adoption, hiring the right talent with the right combination of subject matter expertise and business experience, and the business and technical aspects executives should consider before contemplating the adoption of AI. For more insights on the B2B applications of AI, go to techemergence.com

Avoiding Common Mistakes in Applying AI to Business Problems - with Jeremy Barnes of Element AI

Sales forecasting is big business. If you can better predict how much of a certain product or service you will sell in a given day, you can better stock inventory, better staff your facilities, and ultimately keep more margin in your business's accounts. This week on AI in Industry we interview Dr. John-Paul B Clarke, professor at Georgia Tech and co-founder / Chief Scientist at Pace (previously called "Prix"). Dr. Clarke shares details about how sales predictions are done today, and what AI advancements may allow for in helping businesses sell everything from groceries to hotel rooms. Read the full interview article online at:&nbsp; techemergence.com/ai-sales-forecasting-works-matters

AI for Sales Forecasting - How it Works and Where it Matters

In this O’Reilly Podcast, Rachel Roumeliotis, vice president of content strategy at O’Reilly Media, speaks with Atif Kureishy, global VP of emerging practices and artificial intelligence and deep learning at Teradata, a data and analytics company.  They discuss how enterprises are currently investing in artificial intelligence (AI), which industries are seeing the most impact, barriers to AI implementation, and future trends.

Atif Kureishy on How Enterprises are Investing in AI

In this episode of the Data Show, I spoke with Mark Hammond, founder and CEO of Bonsai, a startup at the forefront of developing AI systems in industrial settings. While many articles have been written about developments in computer vision, speech recognition, and autonomous vehicles, I’m particularly excited about near-term applications of AI to manufacturing, robotics, and industrial automation. In a recent post, I outlined practical applications of reinforcement learning (RL)—a type of machine learning now being used in AI systems. In particular, I described how companies like Bonsai are applying RL to manufacturing and industrial automation. As researchers explore new approaches for solving RL problems, I expect many of the first applications to be in industrial automation.

Machine Learning Needs Machine Teaching

Episode summary: In this week’s episode of AI in Industry, we speak with Larry Lafferty, the President and CEO of Veloxiti. Larry has been building large AI projects for DARPA and other large private companies for the last 30 years. In this interview, Larry explains three critical factors to applying artificial intelligence in the enterprise (with insights especially relevant for companies who aren’t very familiar with AI and data science). AI vendors and business leaders should find the “how to” insights in this interview useful – particularly Larry’s details on organizing data and defining an AI-applicable business problem. Interested readers can listen to the full interview with Larry here:&nbsp;https://www.techemergence.com/how-to-apply-ai-…h-larry-lafferty/

How to Apply AI to an Existing Business with Larry Lafferty

As machine learning models get into the hands of more and more users, there's an increasing expectation that black box isn't good enough: users want to understand why the model made a given prediction, not just what the prediction itself is. This is motivating a lot of work into feature important and model interpretability tools, and one of the most exciting new ones is based on Shapley Values from game theory. In this episode, we'll explain what Shapley Values are and how they make a cool approach to feature importance for machine learning.

Game Theory for Model Interpretability: Shapley Values

Shapley values in machine learning are an interesting and useful enough innovation that we figured hey, why not do a two-parter? Our last episode focused on explaining what Shapley values are: they define a way of assigning credit for outcomes across several contributors, originally to understand how impactful different actors are in building coalitions (hence the game theory background) but now they're being cross-purposed for quantifying feature importance in machine learning models. This episode centers on the computational details that allow Shapley values to be approximated quickly, and a new package called SHAP that makes all this innovation accessible.

SHAP: Shapley Values in Machine Learning

A little ditty from the guts of our iPhone app

Cha-Ching!

In this episode of the Data Show, I spoke with Harish Doddi, co-founder and CEO of Datatron, a startup focused on helping companies deploy and manage machine learning models. As companies move from machine learning prototypes to products and services, tools and best practices for productionizing and managing models are just starting to emerge. Today’s data science and data engineering teams work with a variety of machine learning libraries, data ingestion, and data storage technologies. Risk and compliance considerations mean that the ability to reproduce machine learning workflows is essential to meet audits in certain application domains. And as data science and data engineering teams continue to expand, tools need to enable and facilitate collaboration.

As someone who specializes in helping teams turn machine learning prototypes into production-ready services, I wanted to hear what Doddi has learned while working with organizations that aspire to “become machine learning companies.”

Simplifying Machine Learning Lifecycle Management

AI has made it easier to understand text as a medium in a deeper, more efficient way and at scale. With video, the situation is quite different. Searching for content within videos is more challenging because video is not just voice and sound, it is also a collection of moving and still images on screen. How could AI work to overcome that challenge? In this episode of the AI in Industry podcast, we interview Manish Gupta, CEO and co-founder of VideoKen, about the future of video search as machine learning is increasingly integrated into the process. Dr. Gupta talks about how video is becoming more searchable and discusses his own forecasts about what that will look like in the future. He also predicts what machine learning will allow Youtube to do as people continue to search for more specific video content. Our Content Lead, Raghav Bharadwaj, joins us for this interview. See the full video article here: www.techemergence.com/artificial-intelligence-video-search-youtube-future

Artificial Intelligence for Video Search and the YouTube of the Future

In this episode of the Data Show, I spoke with Sharad Goel, assistant professor at Stanford, and his student Sam Corbett-Davies. They recently wrote a survey paper, “A Critical Review of Fair Machine Learning,” where they carefully examined the standard statistical tools used to check for fairness in machine learning models. It turns out that each of the standard approaches (anti-classification, classification parity, and calibration) has limitations, and their paper is a must-read tour through recent research in designing fair algorithms. We talked about their key findings, and, most importantly, I pressed them to list a few best practices that analysts and industrial data scientists might want to consider.

Why It's Hard to Design Fair Machine Learning Models

How do you think can the concept of 2-millimeter shift impacts you?

If you enjoyed this episode, check out show notes, resources, and more at www.superdatascience.com/198

The Super Data Science Podcast: Big Data | Analytics Careers | Mentors | Success

SDS 198: Two Millimeter Shifts

In this episode of the Data Show, I spoke with Francesca Lazzeri, an AI and machine learning scientist at Microsoft, and her colleague Jaya Mathew, a senior data scientist at Microsoft. We conducted a couple of surveys this year—“How Companies Are Putting AI to Work Through Deep Learning” and “The State of Machine Learning Adoption in the Enterprise”—and we found that while many companies are still in the early stages of machine learning adoption, there’s considerable interest in moving forward with projects in the near future. Lazzeri and Mathew spend a considerable amount of time interacting with companies that are beginning to use machine learning and have experiences that span many different industries and applications. I wanted to learn some of the processes and tools they use when they assist companies in beginning their machine learning journeys.

Lessons Learned While Helping Enterprises Adopt Machine Learning

In this episode of the Data Show, I spoke with Alex Wong, associate professor at the University of Waterloo, and co-founder of DarwinAI, a startup that uses AI to address foundational challenges with deep learning in the enterprise. As the use of machine learning and analytics become more widespread, we’re beginning to see tools that enable data scientists and data engineers to scale and tackle many more problems and maintain more systems. This includes automation tools for the many stages involved in data science, including data preparation, feature engineering, model selection, and hyperparameter tuning, as well as tools for data engineering and data operations.

Wong and his collaborators are building solutions for enterprises, including tools for generating efficient neural networks and for the performance analysis of networks deployed to edge devices.

Tools for Generating Deep Neural Networks With Efficient Network Architectures

For the end-of-year holiday episode of the Data Show, I turned the tables on Data Show host Ben Lorica to talk about trends in big data, machine learning, and AI, and what to look for in 2019. Lorica also showcased some highlights from our upcoming Strata Data and Artificial Intelligence conferences.

Trends in Data, Machine Learning, and AI

In this episode of the Data Show, I spoke with Roger Chen, co-founder and CEO of Computable Labs, a startup focused on building tools for the creation of data networks and data exchanges. Chen has also served as co-chair of O'Reilly's Artificial Intelligence Conference since its inception in 2016. This conversation took place the day after Chen and his collaborators released an interesting new white paper, "Fair value and decentralized governance of data." Current-generation AI and machine learning technologies rely on large amounts of data, and to the extent they can use their large user bases to create “data silos,” large companies in large countries (like the U.S. and China) enjoy a competitive advantage. With that said, we are awash in articles about the dangers posed by these data silos. Privacy and security, disinformation, bias, and a lack of transparency and control are just some of the issues that have plagued the perceived owners of “data monopolies.”

Acquiring and Sharing High-Quality Data

In this episode of the Data Show, I speak with Cassie Kozyrkov, technical director and chief decision scientist at Google Cloud. She describes "decision intelligence" as an interdisciplinary field concerned with all aspects of decision-making, and which combines data science with the behavioral sciences. Most recently she has been focused on developing best practices that can help practitioners make safe, effective use of AI and data. Kozyrkov uses her platform to help data scientists develop skills that will enable them to connect data and AI with their organizations' core businesses.

We had a great conversation spanning many topics, including:

How data science can be more useful

The importance of the human side of data

The leadership talent shortage in data science

Is data science a bubble?

Make Data Science More Useful

In this episode of the Data Show, I speak with Alex Ratner, project lead for Stanford’s Snorkel open source project; Ratner also recently garnered a faculty position at the University of Washington and is currently working on a company supporting and extending the Snorkel project. Snorkel is a framework for building and managing training data. Based on our survey from earlier this year, labeled data remains a key bottleneck for organizations building machine learning applications and services.

Ratner was a guest on the podcast a little over two years ago when Snorkel was a relatively new project. Since then, Snorkel has added more features, expanded into computer vision use cases, and now boasts many users, including Google, Intel, IBM, and other organizations. Along with his thesis advisor professor Chris Ré of Stanford, Ratner and his collaborators have long championed the importance of building tools aimed squarely at helping teams build and manage training data. With today’s release of Snorkel version 0.9, we are a step closer to having a framework that enables the programmatic creation of training data sets.

Labeling, Transforming, and Structuring Training Data Sets for Machine Learning

In this episode of the Data Show, I speak with Kesha Williams, technical instructor at A Cloud Guru, a training company focused on cloud computing. As a full stack web developer, Williams became intrigued by machine learning and started teaching herself the ML tools on Amazon Web Services. Fast forward to today, Williams has built some well-regarded Alexa skills, mastered ML services on AWS, and has now firmly added machine learning to her developer toolkit.

Becoming a Machine Learning Practitioner

In this episode of the Data Show, I speak with Michael Mahoney, a member of RISELab, the International Computer Science Institute, and the Department of Statistics at UC Berkeley. A physicist by training, Mahoney has been at the forefront of many important problems in large-scale data analysis. On the theoretical side, his works spans algorithmic and statistical methods for matrices, graphs, regression, optimization, and related problems. On the applications side, he has contributed to systems used for internet and social media analysis, social network analysis, as well as for a host of applications in the physical and life sciences. Most recently, he has been working on deep neural networks, specifically developing theoretical methods and practical diagnostic tools that should be helpful to practitioners who use deep learning.

Understanding Deep Neural Networks

O'Reilly Media spreads the knowledge of innovators. At O’Reilly, a big part of our business is paying attention to what’s new and interesting in the world of technology. We have a pretty good record at having anticipated some of the big technology developments in recent history. For instance, we launched the first commercial Web site, GNN, in 1993; we organized the meeting at which the term “open source” was first adopted; we were early investors in Blogger, which helped launch the blogging revolution; and more recently, our Web 2.0 conference launched a world-wide meme. We call this predictive sense the “O’Reilly Radar.” And while we’re certainly not always right, we are, at least, good at making interesting guesses.

Our methodology is simple: we draw from the wisdom of the alpha geeks in our midst, paying attention to what’s interesting to them, amplifying these weak signals, and seeing where they fit into the innovation ecology. Add to that the original research conducted by our Research team, and you start to get a good picture of what the technology world is thinking about.

Why It's Hard to Design Fair Machine Learning Models

O'Reilly Radar