Practical AI

Joe Doliner (JD) joined the show to talk about productionizing ML/AI with Pachyderm, an open source data science platform built on Kubernetes (k8s). We talked through the origins of Pachyderm, challenges associated with creating infrastructure for machine learning, and data and model versioning/provenance. He also walked us through a process for going from a Jupyter notebook to a production data pipeline. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – DigitalOcean is simplicity at scale. Whether your business is running one virtual machine or ten thousand, DigitalOcean gets out of your way so your team can build, deploy, and scale faster and more efficiently. New accounts get $100 in credit to use in your first 60 days. Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Featuring:Joe Doliner – Website, GitHub, XChris Benson – Website, GitHub, LinkedIn, XDaniel Whitenack – Website, GitHub, XShow Notes: Pachyderm Pachyderm on GitHub Pachyderm tutorials DoD challenge built using Pachyderm Something missing or broken? PRs welcome!

Fully Connected – a series where Chris and Daniel keep you up to date with everything that’s happening in the AI community. This week we discuss BERT, a new method of pre-training language representations from Google for natural language processing (NLP) tasks. Then we tackle Facebook’s Horizon, the first open source reinforcement learning platform for large-scale products and services. We also address synthetic data, and suggest a few learning resources. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Algolia – Our search partner. Algolia’s full suite search APIs enable teams to develop unique search and discovery experiences across all platforms and devices. We’re using Algolia to power our site search here at Changelog.com. Get started for free and learn more at algolia.com. Featuring:Chris Benson – Website, GitHub, LinkedIn, XDaniel Whitenack – Website, GitHub, XShow Notes:News/Discussion: Is artificial intelligence set to become art’s next medium? BERT (a new method for obtaining rich contextual language representations during pre-training): Google research article TensorFlow BERT BERT paper PyTorch BERT NY Times article Example Colab notebook BERT explained article Transformer paper Google Open Sources BERT to Train Natural Language Models Without Breaking the Bank Horizon: The first open source reinforcement learning platform for large-scale products and services Does Synthetic Data Hold The Secret To Artificial Intelligence? AI Experts: Moving forward with AI likely a series of small steps, not giant leaps Learning resources: The Backpropagation Algorithm Demystified Books “Grokking Deep Learning” by Andrew Trask Something missing or broken? PRs welcome!

We recently met up with Cormac Brick (Intel) and Mike Del Balso (Uber) at O’Reilly AI in SF. As the director of machine intelligence in Intel’s Movidius group, Cormac is an expert in porting deep learning models to all sorts of embedded devices (cameras, robots, drones, etc.). He helped us understand some of the techniques for developing portable networks to maximize performance on different compute architectures. In our discussion with Mike, we talked about the ins and outs of Michelangelo, Uber’s machine learning platform, which he manages. He also described why it was necessary for Uber to build out a machine learning platform and some of the new features they are exploring. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – DigitalOcean is simplicity at scale. Whether your business is running one virtual machine or ten thousand, DigitalOcean gets out of your way so your team can build, deploy, and scale faster and more efficiently. New accounts get $100 in credit to use in your first 60 days. Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Featuring:Cormac Brick – Website, XMike Del Balso – WebsiteChris Benson – Website, GitHub, LinkedIn, XDaniel Whitenack – Website, GitHub, XShow Notes: Intel’s Movidius Group OpenVINO Toolkit ngraph Uber’s Michelangelo Something missing or broken? PRs welcome!

Brett Gaylor joins Chris and Daniel to chat about the recently announced winners of Mozilla’s creative media awards, which focuses on exposing the impact of AI on society. These winners include a film that responds to the audience (via AI recognized emotions) and an interesting chatbot called Wanda. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – DigitalOcean is simplicity at scale. Whether your business is running one virtual machine or ten thousand, DigitalOcean gets out of your way so your team can build, deploy, and scale faster and more efficiently. New accounts get $100 in credit to use in your first 60 days. Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Featuring:Brett Gaylor – Website, XChris Benson – Website, GitHub, LinkedIn, XDaniel Whitenack – Website, GitHub, XShow Notes: Winners of the Creative Media Awards Do not track movie The Science behind Cambridge Analytica Mozilla manifesto Machine bias article from ProPublica Practical AI episode from Lindsey Zuloaga about bias Something missing or broken? PRs welcome!

Himani Agrawal joins Daniel and Chris to talk about how she got into data science and artificial intelligence, and offers advice to others getting into these fields. She goes on to describe the role of artificial intelligence and machine learning within AT&T and telecom in general. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – DigitalOcean is simplicity at scale. Whether your business is running one virtual machine or ten thousand, DigitalOcean gets out of your way so your team can build, deploy, and scale faster and more efficiently. New accounts get $100 in credit to use in your first 60 days. Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Featuring:Himani Agrawal – Website, GitHubChris Benson – Website, GitHub, LinkedIn, XDaniel Whitenack – Website, GitHub, XShow Notes: MLConf SF Grace Hopper Celebration Galvanize Fast.ai Thinkful bootcamp Magenta project Something missing or broken? PRs welcome!

In this new and updates show, Daniel and Chris discuss, among other things, efforts to use AI in art and efforts to make AI interfaces look human. They also discuss some learning resources related to neural nets, AI fairness, and reinforcement learning. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – DigitalOcean is simplicity at scale. Whether your business is running one virtual machine or ten thousand, DigitalOcean gets out of your way so your team can build, deploy, and scale faster and more efficiently. New accounts get $100 in credit to use in your first 60 days. Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Featuring:Chris Benson – Website, GitHub, LinkedIn, XDaniel Whitenack – Website, GitHub, XShow Notes:News from Daniel: AI portrait up for auction Graph ML related things: Deepmind Graphnet library Knowledge Graphs and ML Semantic Scholar CSV conf News from Chris: Magic Leap’s new AI assistant looks alarmingly human MIT Stephen A. Schwarzman College of Computing Deep-learning algorithm identifies dense tissue in mammograms Learning resources: Neural network playground IBM AI Fairness 360 Towards Data Science Artificial Intelligence: What’s The Difference Between Deep Learning And Reinforcement Learning? Something missing or broken? PRs welcome!

Lindsey Zuloaga joins us to discuss bias in hiring, bias in AI, and how we can fight bias in hiring with AI. Lindsey tells us about her experiences fighting bias at HireVue, where she is director of data science, and she gives some practical advice to AI practitioners about fairness in models and data. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – DigitalOcean is simplicity at scale. Whether your business is running one virtual machine or ten thousand, DigitalOcean gets out of your way so your team can build, deploy, and scale faster and more efficiently. New accounts get $100 in credit to use in your first 60 days. Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Featuring:Lindsey Zuloaga – Website, XChris Benson – Website, GitHub, LinkedIn, XDaniel Whitenack – Website, GitHub, XShow Notes: Compas recidivism model HireVue IBM fairness 360 Attacking discrimination with smarter machine learning Something missing or broken? PRs welcome!

Chris and Daniel are back together in another news/updates show. They discuss PyTorch v1.0, some disturbing uses of AI for tracking social credit, and learning resources to get you started with machine learning. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – DigitalOcean is simplicity at scale. Whether your business is running one virtual machine or ten thousand, DigitalOcean gets out of your way so your team can build, deploy, and scale faster and more efficiently. New accounts get $100 in credit to use in your first 60 days. Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Featuring:Chris Benson – Website, GitHub, LinkedIn, XDaniel Whitenack – Website, GitHub, XShow Notes:News from Daniel: Introducing PyTorch across Google cloud VentureBeat article about PyTorch v1.0 Google’s general recommended practices for AI Deep learning made easier with transfer learning Decision tree visualization Anatomy of an AI system It’s hacktoberfest! Conference calls: AMLD O’Reilly AI NY CPVR News from Chris: TensorFlow RoadMap, TensorFlow 2 Leave No Dark Corner The 5 Steps To Build A Business’s Deep Learning Workflow The Pentagon is investing $2 billion into artificial intelligence Google Dataset Search What do you get when you mix sugar water, 3D printing and deep learning? Learning resources: Udacity Machine Learning Course by GA Tech Machine learning cheat sheets Something missing or broken? PRs welcome!

NVIDIA Chief Scientist Bill Dally joins Daniel Whitenack and Chris Benson for an in-depth conversation about ‘everything AI’ at NVIDIA. As the leader of NVIDIA Research, Bill schools us on GPUs, and then goes on to address everything from AI-enabled robots and self-driving vehicles, to new AI research innovations in algorithm development and model architectures. This episode is so packed with information, you may want to listen to it multiple times. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – DigitalOcean is simplicity at scale. Whether your business is running one virtual machine or ten thousand, DigitalOcean gets out of your way so your team can build, deploy, and scale faster and more efficiently. New accounts get $100 in credit to use in your first 60 days. Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Featuring:Bill Dally – WebsiteChris Benson – Website, GitHub, LinkedIn, XDaniel Whitenack – Website, GitHub, XShow Notes: NVIDIA Research NVIDIA Deep Learning Accelerator (NVDLA) CUDA AI-Powered Self-Driving Cars Progressive GANs Optical Flow and Deep Learning TensorRT Deep Learning Super Sampling NVIDIA DGX2 NVSwitch and NVLink NVIDIA Deep Learning Institute Something missing or broken? PRs welcome!

We met up with Wojciech Zaremba at the O’Reilly AI conference in SF. He took some time to talk to us about some of his recent research related to reinforcement learning and robots. We also discussed AI safety and the hype around OpenAI. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:Fastly – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com. Rollbar – We catch our errors before our users do because of Rollbar. Resolve errors in minutes, and deploy your code with confidence. Learn more at rollbar.com/changelog. Linode – Our cloud server of choice. Deploy a fast, efficient, native SSD cloud server for only $5/month. Get 4 months free using the code changelog2018. Start your server - head to linode.com/changelog Featuring:Wojciech Zaremba – Website, GitHub, XDaniel Whitenack – Website, GitHub, XShow Notes: Woj’s talk at O’Reilly AI OpenAI OpenAI blog The Dota playing team of AIs from OpenAI Microsoft Tay Chatbot Reinforcement learning Wikipedia Berkeley course CS294-112 on Reinforcement learning. Includes pdf notes and YouTube videos of the lectures. Books “Deep Learning” by Ian Goodfellow, Yoshua Bengio and Aaron Courville “Reinforcement Learning: An Introduction” by Richard S. Sutton and Andrew G. Barto Something missing or broken? PRs welcome!