Skip to content

📚 Reinforcement Learning — Study Notes

Reinforcement Learning

View the live site — ijk37.com

Notes

Home  |  Exercises  |  Quiz Hub  |  Projects  |  Resources

Beginner-friendly, chapter-wise notes for Reinforcement Learning: An Introduction (2nd edition) by Richard S. Sutton & Andrew G. Barto.

These notes aim to sit in the sweet spot: more explanatory than a cheat sheet, gentler than the textbook. Each note has intuition, worked examples, the essential math (explained, not just stated), and "key takeaways." Read them in order — they build on each other and link forward/back.

Naming: xx-yy-topic.md where xx = chapter number, yy = concept order within the chapter.


🗺️ How to use these notes

  1. Go in order. RL is cumulative — the Bellman equation (3.4) underlies almost everything after it.
  2. Don't skip the worked examples. Tic-tac-toe, the 10-armed testbed, gridworlds, the cliff, Mountain Car — each one makes an abstract idea concrete.
  3. Re-derive the boxed equations by hand once. Especially: \(G_t = R_{t+1} + \gamma G_{t+1}\), the Bellman equations, and the TD update.
  4. Code as you go. The notes flag what to implement; building the algorithms is how it sticks.

Part 0 — Foundations

Chapter 1 · Introduction

Chapter 2 · Multi-armed Bandits


Part I — Tabular Solution Methods

Chapter 3 · Finite Markov Decision Processes

Chapter 4 · Dynamic Programming

Chapter 5 · Monte Carlo Methods

Chapter 6 · Temporal-Difference Learning ⭐

Chapter 7 · n-step Bootstrapping

Chapter 8 · Planning and Learning with Tabular Methods


Part II — Approximate Solution Methods

Chapter 9 · On-policy Prediction with Approximation

Chapter 10 · On-policy Control with Approximation

Chapter 11 · Off-policy Methods with Approximation

Chapter 12 · Eligibility Traces

Chapter 13 · Policy Gradient Methods ⭐


Part III — Looking Deeper

Chapter 14 · Psychology

Chapter 15 · Neuroscience

Chapter 16 · Applications and Case Studies

Chapter 17 · Frontiers


🧭 The mental model to carry through everything

Almost every method in this book is an instance of Generalized Policy Iteration (GPI):

   ┌─────────── evaluation ───────────┐
   │   make the value function agree   │
 policy π  ◄────────────────────────►  value function V/Q
   │   make the policy greedy w.r.t.   │
   └────────── improvement ───────────┘

When you meet any new algorithm, ask: "How does it evaluate? How does it improve? Does it bootstrap? Does it need a model? On-policy or off-policy?" Those five questions place it on the map.


📖 Source

  • Sutton, R. S., & Barto, A. G. (2018). Reinforcement Learning: An Introduction (2nd ed.). MIT Press. Freely available from the authors.

Notes written as a beginner-first study companion. Equations render on GitHub (LaTeX in $…$/$$…$$); diagrams use Mermaid where supported.