Practical AI: Machine Learning, Data Science, LLM cover image

Practical AI: Machine Learning, Data Science, LLM

Latest episodes

undefined
Jun 11, 2019 • 52min

GANs, RL, and transfer learning oh my!

Daniel and Chris explore three potentially confusing topics - generative adversarial networks (GANs), deep reinforcement learning (DRL), and transfer learning. Are these types of neural network architectures? Are they something different? How are they used? Well, If you have ever wondered how AI can be creative, wished you understood how robots get their smarts, or were impressed at how some AI practitioners conquer big challenges quickly, then this is your episode! Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $50 credit. Learn more at do.co/changelog. Discover.bot – A digital space for bot developers and enthusiasts of all skill levels to learn from one another, share stories, and move the bot conversation forward. Want to learn more about building bots? Get started with their Guide to Bot Building Frameworks. GitPrime – Download GitPrime’s 20 Patterns book, a field guide to help engineering managers recognize achievement, spot bottlenecks, and debug development processes with data. 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. Featuring:Chris Benson – Twitter, GitHub, LinkedIn, WebsiteDaniel Whitenack – Twitter, GitHub, WebsiteShow Notes:SUBSCRIBE ~> Brain Science — For the curious! We’re exploring the inner-workings of the human brain to understand behavior change, habit formation, mental health, and being human. RL (Reinforcement learning) Good overview article Human-level performance in Quake III–Capture the Flag Practical AI episode about Deep Reinforcement Learning Practical AI episode about OpenAI, reinforcement learning, robots, and safety PyTorch RL tutorial GANs (Generative Adversarial Networks) Good overview article This is not a person website Christie’s AI art auction OpenAI Generative models TensorFlow GAN tutorial Transfer learning Good overview article Forbes article on Google AutoML Practical AI episode on BERT Practical AI episode on GPT-2 How to build a State-of-the-Art Conversational AI with Transfer Learning -NAACL workshop on transfer learning for NLP Something missing or broken? PRs welcome!
undefined
Jun 4, 2019 • 46min

Visualizing and understanding RNNs

Andreas Madsen, a freelance ML/AI engineer and Distill.pub author, joins us to discuss his work visualizing neural networks and recurrent neural units. Andreas discusses various neural unites, RNNs in general, and the “why” of neural network visualization. He also gives us his perspective on ML/AI freelancing and moving from web development to AI research. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $50 credit. Learn more at do.co/changelog. Discover.bot – A digital space for bot developers and enthusiasts of all skill levels to learn from one another, share stories, and move the bot conversation forward. Want to learn more about building bots? Get started with their Guide to Bot Building Frameworks. strongDM – Manage access to any database, server, and environment. strongDM makes it easy for DevOps to enforce the controls InfoSec teams require. 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. Featuring:Andreas Madsen – Twitter, GitHubChris Benson – Twitter, GitHub, LinkedIn, WebsiteDaniel Whitenack – Twitter, GitHub, WebsiteShow Notes: Distill.pub by Andreas Taylor approximation and other mathematics related to neural networks Books “Programming Collective Intelligence” by Toby Segaran Something missing or broken? PRs welcome!
undefined
May 28, 2019 • 1h 2min

How to get plugged into the AI community

Chris and Daniel take you on a tour of local and global AI events, and discuss how to get the most out of your experiences. From access to experts to developing new industry relationships, learn how to get your foot in the door and make connections that help you grow as an AI practitioner. Then drawing from their own wealth of experience as speakers, they dive into what it takes to give a memorable world-class talk that your audience will love. They break down how to select the topic, write the abstract, put the presentation together, and deliver the narrative with impact! Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $50 credit. Learn more at do.co/changelog. strongDM – Manage access to any database, server, and environment. strongDM makes it easy for DevOps to enforce the controls InfoSec teams require. Discover.bot – A digital space for bot developers and enthusiasts of all skill levels to learn from one another, share stories, and move the bot conversation forward. Want to learn more about building bots? Get started with their Guide to Bot Building Frameworks. GitPrime – Download GitPrime’s 20 Patterns book, a field guide to help engineering managers recognize achievement, spot bottlenecks, and debug development processes with data. Featuring:Chris Benson – Twitter, GitHub, LinkedIn, WebsiteDaniel Whitenack – Twitter, GitHub, WebsiteShow Notes: Local Meetups - Great for first steps, local help and relationship building, job connections, involvement in local projects, etc. Mostly free! PyData TensorFlow PyTorch Deep Learning etc. Large industry conferences - Great to see what some of the larger companies and innovative startups are doing with AI, application and infra focused, etc. Expensive. O’Reilly AI, Strata Data, TensorFlow World QCon AI, QCon Spark Summit MLConf Applied Machine Learning Days PyCon LiveWorx etc. Smaller, topical events or events focused on particular tools - Great if you are heavily invested in particular tech or a topic like NLP. In general you have to look around quite a bit to find these. TensorFlow dev summit Spacy IRL PyTorch developers conference AllenNLP Summit Applied Artificial Intelligence Conference DataSciConhttps://datascicon.tech/ Etc. Research conferences - Meant to be an outlet for original research. Great to see how research orgs/institutions are pushing the boundaries and doing groundbreaking things. Hard to get into and very academic. NeurIPS EMNLP ICLR CVPR etc. Some conference talks from the Practical AI hosts: Chris Benson’s upcoming talk at LiveWorx 2019 in Boston on Wednesday, June 12, 2019: Artificial Intelligence Strategy: Digital Transformation Through Deep Learning Examples of previous AI talks by Chris Benson Building GPU-Accelerated Workflows with TensorFlow and Kubernetes Fully Reproducible ML Deployment with Spark, Pachyderm, and MLeap Join the community — it’s free! Connect with us in the #practicalai Slack channel. The community is here for you as a place to below, to bounce ideas around, or to get feedback on a conference to attend or talk you’d like to give. Something missing or broken? PRs welcome!
undefined
May 21, 2019 • 57min

AI adoption in the enterprise

At the recent O’Reilly AI Conference in New York City, Chris met up with O’Reilly Chief Data Scientist Ben Lorica, the Program Chair for Strata Data, the AI Conference, and TensorFlow World. O’Reilly’s ‘AI Adoption in the Enterprise’ report had just been released, so naturally Ben and Chris wanted to do a deep dive into enterprise AI adoption to discuss strategy, execution, and implications. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $50 credit. Learn more at do.co/changelog. strongDM – Manage access to any database, server, and environment. strongDM makes it easy for DevOps to enforce the controls InfoSec teams require. Discover.bot – A digital space for bot developers and enthusiasts of all skill levels to learn from one another, share stories, and move the bot conversation forward. Want to learn more about building bots? Get started with their Guide to Bot Building Frameworks. GitPrime – Download GitPrime’s 20 Patterns book, a field guide to help engineering managers recognize achievement, spot bottlenecks, and debug development processes with data. Featuring:Ben Lorica – Twitter, LinkedIn, WebsiteChris Benson – Twitter, GitHub, LinkedIn, WebsiteShow Notes: O’Reilly’s AI Adoption in the Enterprise: How Companies Are Planning and Prioritizing AI Projects in Practice O’Reilly AI Conference 2019 | NYC Something missing or broken? PRs welcome!
undefined
May 14, 2019 • 1h 2min

When AI meets quantum mechanics

Can AI help quantum physicists? Can quantum physicists help the AI community? The answers are yes and yes! Dr. Shohini Ghose from Wilfrid Laurier University and Marcus Edwards from the University of Waterloo join us to discuss ML/AI’s impact on physics and quantum computing potential for ML/AI. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $100 credit. Learn more at do.co/changelog. Discover.bot – A digital space for bot developers and enthusiasts of all skill levels to learn from one another, share stories, and move the bot conversation forward. Want to learn more about building bots? Get started with their Guide to Bot Building Frameworks. strongDM – Manage access to any database, server, and environment. strongDM makes it easy for DevOps to enforce the controls InfoSec teams require. 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. Featuring:Shohini Ghose – WebsiteMarcus Edwards – Twitter, GitHub, LinkedIn, WebsiteChris Benson – Twitter, GitHub, LinkedIn, WebsiteDaniel Whitenack – Twitter, GitHub, WebsiteShow Notes: Quantum Emulation project (email marcus@quantumemulation.com/) Efforts to use Quantum Computing to accelerate ML/AI: from Google from IBM Xanadu: a dedicated ML platform for quantum computers OpenQASM Rigetti Quantum open source foundation TED talk from Dr. Ghose Something missing or broken? PRs welcome!
undefined
May 7, 2019 • 59min

TensorFlow Dev Summit 2019

This week Daniel and Chris discuss the announcements made recently at TensorFlow Dev Summit 2019. They kick it off with the alpha release of TensorFlow 2.0, which features eager execution and an improved user experience through Keras, which has been integrated into TensorFlow itself. They round out the list with TensorFlow Datasets, TensorFlow Addons, TensorFlow Extended (TFX), and the upcoming inaugural O’Reilly TensorFlow World conference. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $100 credit. Learn more at do.co/changelog. Discover.bot – A digital space for bot developers and enthusiasts of all skill levels to learn from one another, share stories, and move the bot conversation forward. Want to learn more about building bots? Get started with their Guide to Bot Building Frameworks. GitPrime – Download GitPrime’s 20 Patterns book, a field guide to help engineering managers recognize achievement, spot bottlenecks, and debug development processes with data. 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. Featuring:Chris Benson – Twitter, GitHub, LinkedIn, WebsiteDaniel Whitenack – Twitter, GitHub, WebsiteShow Notes: TensorFlow Dev Summit 2019 TensorFlow 2.0 alpha release TensorFlow Datasets TensorFlow Extended (TFX) TensorFlow Addons O’Reilly TensorFlow World conference Something missing or broken? PRs welcome!
undefined
Apr 30, 2019 • 1h 3min

CTRL-labs lets you control machines with your mind

No, this isn’t science fiction! CTRL-labs is using neural signals and AI to build neural interfaces. Adam Berenzweig, from CTRL-labs R&D, joins us to explain how this works and how they have made it practical. Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $100 credit. Learn more at do.co/changelog. O'Reilly Open Source Software Conference – OSCON has been ground zero for the open source community for 20 years. This year they’ve expanded to become a “software development conference” — because in 2019, software development IS open source. The program covers everything from open source, AI, infrastructure, blockchain, edge computing, architecture, and emerging languages. Use the code CHANGELOG20 to get 20% off Bronze, Silver, and Gold passes. 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. GitPrime – Download GitPrime’s 20 Patterns book, a field guide to help engineering managers recognize achievement, spot bottlenecks, and debug development processes with data. Featuring:Adam Berenzweig – Twitter, LinkedIn, WebsiteChris Benson – Twitter, GitHub, LinkedIn, WebsiteDaniel Whitenack – Twitter, GitHub, WebsiteShow Notes: O’Reilly AI keynote from CTRL-labs CTRL-labs website CTRL-kit TED talk “The real reason for brains” IBM healthcare Practical AI episode Something missing or broken? PRs welcome!
undefined
Apr 23, 2019 • 46min

Deep Reinforcement Learning

While attending the NVIDIA GPU Technology Conference in Silicon Valley, Chris met up with Adam Stooke, a speaker and PhD student at UC Berkeley who is doing groundbreaking work in large-scale deep reinforcement learning and robotics. Adam took Chris on a tour of deep reinforcement learning - explaining what it is, how it works, and why it’s one of the hottest technologies in artificial intelligence! Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $100 credit. Learn more at do.co/changelog. O'Reilly Open Source Software Conference – OSCON has been ground zero for the open source community for 20 years. This year they’ve expanded to become a “software development conference” — because in 2019, software development IS open source. The program covers everything from open source, AI, infrastructure, blockchain, edge computing, architecture, and emerging languages. Use the code CHANGELOG20 to get 20% off Bronze, Silver, and Gold passes. 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. GitPrime – Download GitPrime’s 20 Patterns book, a field guide to help engineering managers recognize achievement, spot bottlenecks, and debug development processes with data. Featuring:Adam Stooke – Twitter, GitHub, LinkedInChris Benson – Twitter, GitHub, LinkedIn, WebsiteShow Notes: NVIDIA GTC: Doing More with More: Recent Achievements in Large-Scale Deep Reinforcement Learning Accelerated Methods for Deep Reinforcement Learning Accelerated Methods for Deep Reinforcement Learning - code OpenAI Five - Dota 2 AlphaStar: Mastering the Real-Time Strategy Game StarCraft II NVIDIA Graduate Fellowship Crunchbase The Hertz Foundation Something missing or broken? PRs welcome!
undefined
Apr 15, 2019 • 52min

Making the world a better place at the AI for Good Foundation

Longtime listeners know that we’re always advocating for ‘AI for good’, but this week we have taken it to a whole new level. We had the privilege of chatting with James Hodson, Director of the AI for Good Foundation, about ways they have used artificial intelligence to positively-impact the world - from food production to climate change. James inspired us to find our own ways to use AI for good, and we challenge our listeners to get out there and do some good! Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $100 credit. Learn more at do.co/changelog. O'Reilly Open Source Software Conference – OSCON has been ground zero for the open source community for 20 years. This year they’ve expanded to become a “software development conference” — because in 2019, software development IS open source. The program covers everything from open source, AI, infrastructure, blockchain, edge computing, architecture, and emerging languages. Use the code CHANGELOG20 to get 20% off Bronze, Silver, and Gold passes. 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 move fast and fix things because of Rollbar. Resolve errors in minutes. Deploy with confidence. Learn more at rollbar.com/changelog. Featuring:James Hodson – Twitter, WebsiteChris Benson – Twitter, GitHub, LinkedIn, WebsiteDaniel Whitenack – Twitter, GitHub, WebsiteShow Notes: AI For Good Foundation AI For Good Foundation Video Lectures KDD Earth Day “Fragile Earth” Workshop Discounted Membership for Practical AI Podcast Listeners! AI Experts: Moving forward with AI likely a series of small steps, not giant leaps Something missing or broken? PRs welcome!
undefined
Apr 8, 2019 • 49min

GIPHY's celebrity detector

GIPHY’s head of R&D, Nick Hasty, joins us to discuss their recently released celebrity detector project. He gives us all of the details about that project, but he also tells us about GIPHY’s origins, AI in general at GIPHY, and more! Join the discussionChangelog++ members support our work, get closer to the metal, and make the ads disappear. Join today!Sponsors:DigitalOcean – Check out DigitalOcean’s dedicated vCPU Droplets with dedicated vCPU threads. Get started for free with a $100 credit. Learn more at do.co/changelog. Discover.bot – A digital space for bot developers and enthusiasts of all skill levels to learn from one another, share stories, and move the bot conversation forward. Want to learn more about building bots? Get started with their Guide to Bot Building Frameworks. 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. GitPrime – Download GitPrime’s 20 Patterns book, a field guide to help engineering managers recognize achievement, spot bottlenecks, and debug development processes with data. Featuring:Nick Hasty – Twitter, LinkedIn, WebsiteChris Benson – Twitter, GitHub, LinkedIn, WebsiteDaniel Whitenack – Twitter, GitHub, WebsiteShow Notes: Blog post about the celeb detector project GIPHY Enters the Open Source Community with Celebrity Detector (readwrite article) MTCNN method used in the celebrity detector project GitHub repo Demo page 3D projection of celeb face embeddings Complete list of all celebs in the model Something missing or broken? PRs welcome!

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode