user avatar
The TWIML AI Podcast
@twimlai
This Week in #MachineLearning & #AI (podcast) brings you the most interesting and important stories from the world of #ML and artificial intelligence.
Joined May 2016
Posts
  • Pinned
    user avatar
    As context windows grow into the millions of tokens, many AI practitioners are questioning whether retrieval-augmented generation (RAG) is still necessary. If modern models can ingest entire libraries of documents, why bother with retrieval at all? In this episode, Alex Bowcut,
    00:00
  • user avatar
    Today we're joined by @goodfellow_ian of @GoogleBrain @googleresearch and Sandy Huang, PhD Student at @UCBerkeley, to discuss their paper "Adversarial Attacks on Neural Network Policies." Follow the link to check it out! buff.ly/2pd8MNM
  • user avatar
    Policy reinforcement learning in one slide. The math is actually similar to supervised learning. @karpathy @OpenAI #reworkDL #neuralnets
  • user avatar
    Today we kick off our AI Rewind series with @jeremyphoward of @fastdotai. We discuss trends in Deep Learning in 2018 and beyond, including papers, tools and techniques that have contributed to making deep learning more accessible than ever. twimlai.com/talk/214
  • user avatar
    Today on the pod, @math_rachel joins @samcharrington discuss "Practical Deep Learning!" Rachel gives us the rundown on @fastdotai, including the philosophy and goals behind the courses, switching from @TensorFlow to @PyTorch & more! Head over to buff.ly/2Ik3yZ3 to listen!
  • user avatar
    “Benchmarks are not there for us to beat them to death and burn TPU hours… Benchmarks are there for us to test ideas quickly, see if the ideas work, and then move beyond that.” - @georgiagkioxari of @MetaAI
  • user avatar
    Taking a page out of the @valohaiai book, we made our own version of this meme 😏 check out our latest episode to hear the conversation that inspired it!
    Truly groundbreaking. Never to be seen again. #DataScience #MachineLearning #mlops
  • user avatar
    Today, we're joined by @svlevine, associate professor at @Berkeley_EECS and co-founder of @physical_int to discuss π0 (pi-zero), a general-purpose robotic foundation model. We dig into the model architecture, which pairs a vision language model (VLM) with a diffusion-based action
    00:00
  • user avatar
    Today we’re joined by @pabbeel of @Berkeley_EECS, @berkeley_ai, and @CovariantAI, to discuss real-world, industrial applications of AI, his work at the intersection of unsupervised learning and RL, his new podcast @therobotbrains, and much more! twimlai.com/reinforcement-…
  • user avatar
    Today we close out our AI Rewind series joined by Michael Bronstein (@mmbronstein), of @imperialcollege London, and @Twitter @TwitterResearch to discuss Graph Machine Learning in 2020, including uses across domains like physics and bioinformatics.
  • user avatar
    Today we’re excited to be joined by Michael Bronstein (@mmbronstein), Head of Graph Machine Learning at @twitter @TwitterResearch. In our conversation, we discuss the evolution of the graph machine learning space, his new role at Twitter, and much more! twimlai.com/twiml-talk-394…
  • user avatar
    Today we're joined by @seb_ruder of @nuigalway and @_aylien, to discuss trends in Natural Language Processing in 2018 and beyond! We cover a bunch of interesting papers and talk through Sebastian’s predictions for the new year. Happy New Year! twimlai.com/talk/216
  • user avatar
    Today we’re joined by Georgia Gkioxari, research scientist at @FacebookAI Research, to discuss @PyTorch 3D. Georgia walks us through the user experience of PyTorch3D, describing how it fits in the broad goal of giving computers better means of twimlai.com/pytorch-3d-dee…
  • user avatar
    Prioritizing understanding over prediction is at the core of how AI ethics should be reframed. Learn more in our interview with PhD student @Abebab who recently won best paper at the @black_in_ai workshop @NeurIPSConf. buff.ly/2vsnXtc