The Little Book of Reinforcement Learning: A Concise Guide with Pytorch Code
I created this short introduction to Reinforcement Learning to guide readers from fundamental concepts to applied algorithms. Alongside the book, I have shared Pytorch implementations ranging from Monte Carlo methods to PPO, plus rigorous proofs for dynamic programming. Written in 2021 and available under a Creative Commons license, this resource aims to make complex AI topics accessible for everyone.
This book is a short introduction to Reinforcement Learning, from the basics to applied algorithms.
- verdverm
This looks like a good pre-read for Nathan Lambert's https://rlhfbook.com/
- newsomix9xl
Real biological operant behavior isn't exactly trial and error learning.
Many factors shape and guide initial responses.
What I've noticed in some descriptions of models is the use of optimization for reinforcement to shape responses. In real organisms behavior may be controlled by short or long term outcomes, and may oscillate between this "optimization" based on schedules. This produces variability in the trials which can adjust behavior. Are we seeing these reinforcement models do this?
- programjames
I skimmed through the book, and it's lacking the information theory foundations. For example, "trust region methods" come from maximizing the policy's relative entropy (to a reference policy) under a tournament system where high-scoring agents are exponentially likely to survive. In general, a reward is the negative bits it costs an environment to propagate an agent (multiplied by some temperature).