Schedule
The schedule is tentative and subject to change. We will have guest lectures and may accomodate the schedule accordingly.
| Date | Topic | Files and Reading |
|---|---|---|
| Mar 26 | Course overview | Slides, Book Ch. 1 |
| Mar 28 | Intro to Markov Decision Processes | Slides, Book Ch. 2,3 |
| Apr 2 | Planning with a known model in the tabular case | Slides, Book Ch. 4 |
| Apr 4 | Policy Iteration (contd) and MuJoCo setup | Slides, Book Ch. 4 |
| Apr 9 | Policy gradient methods - I | Slides, policy gradient |
| Apr 11 | Policy gradient methods - II | Slides, policy gradient |
| Apr 16 | Off-policy learning - I | |
| Apr 18 | Off-policy learning - II | |
| Apr 23 | Imitation learning | |
| Apr 25 | MCTS and UC Trees | Slides |
| Apr 30 | Trajectory Optimization | Slides |
| May 2 | Combining Trajectories and Policies | Slides |
| May 7 | Guest lecture - Prof. Emanuel Todorov (UW) | Slides |
| May 9 | Guest lecture - Dr. Igor Mordatch (OpenAI) | |
| May 14 | Guest lecture - Dr. Vikash Kumar (Google Brain) | Slides |
| May 16 | No class | |
| May 21 | Learning to learn, meta learning | |
| May 23 | General duality between control & inference, compositionality in LDMPs | |
| May 28 | No class | |
| May 30 | Hierarchical RL | Slides |