📦 Reinforcement Learning — Resources¶
Reference material that backs the notes, exercises, and projects.
The Textbook¶
Everything here is a companion to:
Sutton, R. S., & Barto, A. G. (2018). Reinforcement Learning: An Introduction (2nd ed.). MIT Press.
The book PDF (RL_BartoSutton.pdf) lives in this folder in the repository. It is freely available from the authors, so it is kept local-only (git-ignored) rather than committed here — download your own copy from the official page below.
| Resource | Where |
|---|---|
| Official book page & free PDF | incompleteideas.net/book/the-book-2nd.html |
| Authors' code (Python) | github.com/ShangtongZhang/reinforcement-learning-an-introduction |
| Errata & solutions discussion | linked from the official book page |
How the Layers Map to the Book¶
| This repo | Book coverage |
|---|---|
| Notes | Chapters 1–17, one page per major concept |
| Exercises | Chapter-by-chapter problem sets with worked answers |
| Projects | The book's flagship examples, built from scratch in numpy |
Prerequisites¶
| Tool | Used for |
|---|---|
| Python 3.9+ | Running the projects |
| numpy | The only hard dependency (04-projects/requirements.txt) |
| matplotlib | Optional — scripts print results and only plot if it is installed |
| A little calculus & probability | Following the derivations in the notes |
05-resources/intentionally keeps large/copyrighted files out of git via.gitignore; only this page is published to the site.