Introduction to Reinforcement Learning - Distributed RL Systems - Lecture 9

Introduction to Reinforcement Learning - Distributed RL Systems - Lecture 9

Bolei Zhou via YouTube Direct link

Case Study: AlexNet

15 of 34

15 of 34

Case Study: AlexNet

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Introduction to Reinforcement Learning - Distributed RL Systems - Lecture 9

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Intro
  2. 2 Today's Outline
  3. 3 System and architecture are the foundation
  4. 4 Properties of Distributed Systems
  5. 5 MIT EECS 6.824 Distributed Systems
  6. 6 Updating Model Parameters
  7. 7 Synchronous Update versus Asynchronous Update
  8. 8 Decentralized Asynchronous Stochastic Gradient Descend
  9. 9 Parallelism in Distributed ML Systems
  10. 10 Hogwild: Lock-free asynchronous SGD
  11. 11 Implementation of Hogwild (asych SGD) in PyTorch
  12. 12 Case Study: MapReduce
  13. 13 Case Study: DisBelief
  14. 14 Fun facts about Jeff Dean
  15. 15 Case Study: AlexNet
  16. 16 Diagram of Reinforcement Learning
  17. 17 Development of Distributed RL Systems
  18. 18 2013: Deep Q Network
  19. 19 2015: General Reinforcement Learning Architecture (GORILA)
  20. 20 Review on Actor-Critic Methods
  21. 21 A3C: Asynchronous Advantage Actor Critic (ABC)
  22. 22 Comparison to Variants of DQN and GORILA
  23. 23 Sample code for A3C
  24. 24 Why Asynchronism works in A3C?
  25. 25 Comparison of A3C and A2C
  26. 26 Sample code for A2C
  27. 27 2018: Apex-X (Distributed Prioritized Experience Replay)
  28. 28 2018: IMPALA (Importance Weighted Actor- Learner Architecture)
  29. 29 2018 RLlib: abstraction for distributed RL
  30. 30 Some Other Parallelizable Algorithms: (Revisited) Evolution Strategies
  31. 31 Case Study: Al for Modern Games
  32. 32 System Design for AlphaGo Zero
  33. 33 System Design for AlphaStar
  34. 34 Conclusion

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.