MAP670C - Reinforcement Learning (M2DS) (2023-2024)

Syllabus: This 20-hour course provides an introduction to reinforcement learning. It is based on the new edition of the book "Reinforcement Learning: An Introduction" by R. Sutton and A. Barto. Barto (available online at http://incompleteideas.net/book/the-book-2nd.html).

Outline:

1. Introduction to reinforcement learning and Markov decision processes
2. The bandit case
3. Tabular methods: prediction by dynamic programming, Monte Carlo method and TD Learning
4. Planning and learning for tabular methods
5. approximate methods: prediction, planning and learning

Grading: Project based on a research article