Castuner

Edan Meyer

Edan Meyer

Edan Meyer talks about AI, machine learning and especially reinforcement learning, with educational episodes focused on interesting new research. Links: Personal Site: https://www.edanmeyer.com/ Twitter: https://twitter.com/ejmejm1

Lyssna i din app

Avsnitt

61 avsnitt

  1. Training RL From YouTube Videos 39:32
  2. Chinchilla Explained: Compute-Optimal Massive Language Models 40:30
  3. Neural Networks From Zero: Custom Data 18:33
  4. GPT-4 Outperforms RL by Studying and Reasoning... 🤔 26:31
  5. Neural Networks From Zero: Backpropagation and Testing 17:57
  6. Google's New Model Learns College-Level Math 49:12
  7. I Talked with Rich Sutton 1:38:06
  8. Neural Networks From Zero: Gradient Descent 14:19
  9. Q Learning Tutorial: Testing and Visualizing 18:52
  10. How I Got My First Machine Learning Job (Without a Degree) 18:01
  11. Why Computers Won't Make Themselves Smarter | The Singularity 27:02
  12. Is AI Research On the Wrong Path? 13:38
  13. Reinforcement Learning Made Simple - Q-Values 22:30
  14. Reinforcement Learning Made Simple - Policy 19:27
  15. Reinforcement Learning Made Simple - The Goal 19:18
  16. This is What Limits Current LLMs 10:49
  17. Neural Networks From Zero: Forward Propogation 25:59
  18. Doing ML Research as a Graduate Student 23:03
  19. 2 Years of My Research Explained in 13 Minutes 18:35
  20. This AI Learns from YouTube! 27:34
  21. Neural Networks From Zero: How It Works 13:33
  22. The AI Scaling Problem 20:52
  23. 12 Steps to AGI 52:10
  24. Q Learning Tutorial: Training and Loss Function 29:11
  25. AI Learns to Kill Cows in Runescape 14:11
  26. What's New In Machine Learning? 50:57
  27. Reinforcement Learning Made Simple - Reward 21:39
  28. Learning Forever, Backprop Is Insufficient 34:13
  29. AGI is NOT coming soon 20:59
  30. Q Learning Tutorial: Training Loop 27:52
  31. AlphaCode Explained: AI Code Generation 40:37
  32. Agent Learns to do Reinforcement Learning 32:30
  33. Q Learning Tutorial: Experience Replay 34:43
  34. The Bellman Equations Explained - RL Theory 18:44
  35. The Best of NeurIPS 2022 32:24
  36. Q Learning Tutorial: Deep Q Networks 23:02
  37. Using ChatGPT to Write My PhD Apps 23:24
  38. Q Learning AI Explained 7:57
  39. Is Gato Really the Future of AI? 29:02
  40. Model Based RL Finally Works! 35:45
  41. Making an AI to Answer Our Questions! (Part 2) 21:35
  42. Stable Diffusion - What, Why, How? 1:01:52
  43. This Algorithm Could Make a GPT-4 Toaster Possible 47:06
  44. Making an AI to Answer Our Questions! (Part 1) 23:11
  45. RL Foundation Models Are Coming! 37:48
  46. Reinforcement Learning: Training and Testing 28:37
  47. ChatGPT Is a Dead End 16:30
  48. Reinforcement Learning: Deep Policy Network 16:31
  49. DALL-E 2 is… meh 22:47
  50. GPT-4: What, Why, How? 43:56