shangtongzhang/reinforcement-learning-an-introduction

This repository is a Python recreation of the code examples and figures from a textbook called Reinforcement Learning: An Introduction (Second Edition) by Richard Sutton and Andrew Barto. That book is a well known starting point for the field of reinforcement learning, which is a branch of machine learning where an agent learns to make decisions by trying things and seeing what rewards or penalties follow. The repository does not teach reinforcement learning from scratch on its own, it assumes you have the book open beside you and want runnable code for the examples and plots inside it. The README is almost entirely a table of contents organised chapter by chapter, matching the structure of the book. For each chapter it lists the figures and worked examples that have been reproduced, with links to the resulting plot images stored in the repository. The covered chapters include Tic Tac Toe in Chapter 1, multi armed bandits in Chapter 2, gridworld dynamic programming in Chapters 3 and 4, Monte Carlo methods and blackjack in Chapter 5, temporal difference learning with Sarsa, Q learning, and cliff walking in Chapter 6, n step methods in Chapter 7, Dyna and planning in Chapter 8, function approximation in Chapter 9, and the Mountain Car task in Chapter 10. There is a short note in the README asking readers to open GitHub issues rather than emailing the author if they hit bugs or confusion in the code, and stating that the repository does not include solutions to the book's exercises. There are no installation instructions, no usage examples, and no description of dependencies in the part of the README that was shown. It functions more as an index that maps each book figure to a script and an image rather than as a guided tutorial.