【增强学习】CS294-112深度增强学习课程（加州大学伯克利分校 2017）（部分英文字幕）
https://www.youtube.com/playlist?list=PLkFD6_40KJIwTmSbCv9OVJB3YaO4sFwkX
CS294-112 Deep Reinforcement Learning Sp17 课程主页：http://rll.berkeley.edu/deeprlcourse/
视频选集
（1/21）
1.18 Introduction and course overview (Levine, Finn, Schulman)
01:18:27
1.23 Supervised learning and decision making (Levine)
01:12:12
1.25 Optimal control and planning (Levine)
01:03:11
1.30 Learning dynamical system models from data (Levine)
01:22:14
2.1 Learning policies by imitating optimal controllers (Levine)
01:09:09
2.6 Guest lecture: Igor Mordatch, OpenAI
01:20:13
2.8 RL definitions, value iteration, policy iteration (Schulman)
51:52
2.13 Reinforcement learning with policy gradients (Schulman)
01:05:19
2.15 Learning Q-functions: Q-learning, SARSA, and others (Schulman)
01:17:23
2.22 Advanced Q-learning: replay buffers, target networks, double Q-learning (Sc
01:20:16
2.27
01:16:33
3.1 Advanced topics in imitation and safety (Finn)
01:23:33
3.6 Inverse RL: acquiring objectives from demonstration (Finn)
01:14:16
3.8 Advanced policy gradients: natural gradient and TRPO (Schulman)
01:24:11
3.13 Policy gradient variance reduction and actor-critic algorithms (Schulman)
01:20:38
3.15 Summary of policy gradients and temporal difference methods (Schulman)
01:12:12
3.20 The exploration problem (Schulman)
01:21:47
3.22 Parallel RL algorithms, open problems and challenges in deep reinforcement
01:18:35
4.3 Transfer in Reinforcement Learning (Finn)
01:24:46
4.5 Neural Architecture Search with Reinforcement Learning: Quoc Le and Barret Z
01:16:07
4.10 Generalization and Safety in Reinforcement Learning and Control: Aviv Tamar
01:16:50
顶部