← Books Explained

// BOOK COMPANION · 17/17 LIVE

Reinforcement Learning: An Introduction — explained.

An easy-to-read, detailed companion to Richard S. Sutton and Andrew G. Barto, Reinforcement Learning: An Introduction, 2nd edition. Each chapter builds intuition first, then explains the equation, works through an example, warns about the common mistake, answers practical questions, and ends with a compact study card.

Independent learning material, written in original language. It follows the official chapter sequence without copying the book. All 17 chapters are complete and ready to read.


CHAPTERS · 17 LIVE

chapter 1 · Foundations

Introduction live

The reinforcement-learning problem, its four main elements, delayed consequences, and why trial-and-error learning is different from supervised learning.

plain-language notes· study card
chapter 2 · Part I · Tabular Methods

Multi-armed Bandits live

The exploration–exploitation dilemma in its cleanest form, including ε-greedy action selection, optimistic values, UCB, and gradient bandits.

plain-language notes· study card
chapter 3 · Part I · Tabular Methods

Finite Markov Decision Processes live

The mathematical language of sequential decisions: states, actions, rewards, returns, policies, value functions, and Bellman equations.

plain-language notes· study card
chapter 4 · Part I · Tabular Methods

Dynamic Programming live

How a known environment model enables policy evaluation, policy improvement, policy iteration, value iteration, and generalized policy iteration.

plain-language notes· study card
chapter 5 · Part I · Tabular Methods

Monte Carlo Methods live

Learning values and policies from complete sampled episodes, with on-policy control and off-policy importance sampling.

plain-language notes· study card
chapter 6 · Part I · Tabular Methods

Temporal-Difference Learning live

Learning before an episode ends by bootstrapping, then extending TD prediction to Sarsa, Q-learning, Expected Sarsa, and Double Q-learning.

plain-language notes· study card
chapter 7 · Part I · Tabular Methods

n-step Bootstrapping live

The continuum between one-step TD and Monte Carlo, including n-step Sarsa, off-policy correction, Tree Backup, and Q(σ).

plain-language notes· study card
chapter 8 · Part I · Tabular Methods

Planning and Learning with Tabular Methods live

How learned models turn experience into simulated experience through Dyna, prioritized sweeping, rollout planning, and Monte Carlo tree search.

plain-language notes· study card
chapter 9 · Part II · Approximation

On-policy Prediction with Approximation live

Replacing value tables with parameterized functions, defining a prediction objective, and learning with linear features, tile coding, and neural networks.

plain-language notes· study card
chapter 10 · Part II · Approximation

On-policy Control with Approximation live

Extending approximate value functions to action selection, episodic semi-gradient Sarsa, and average-reward control for continuing tasks.

plain-language notes· study card
chapter 11 · Part II · Approximation

Off-policy Methods with Approximation live

Why bootstrapping, function approximation, and off-policy learning can diverge—and how Gradient-TD and emphatic methods restore stability.

plain-language notes· study card
chapter 12 · Part II · Approximation

Eligibility Traces live

A memory of recently active features that efficiently blends many n-step returns through TD(λ), Sarsa(λ), and true-online variants.

plain-language notes· study card
chapter 13 · Part II · Approximation

Policy Gradient Methods live

Directly parameterizing the policy, deriving the policy-gradient idea, reducing variance with baselines, and combining actors with critics.

plain-language notes· study card
chapter 14 · Part III · Looking Deeper

Psychology live

Connections between computational RL and animal learning: classical and instrumental conditioning, prediction error, habits, and cognitive maps.

plain-language notes· study card
chapter 15 · Part III · Looking Deeper

Neuroscience live

How reward-prediction errors relate to dopamine signals, what supports the correspondence, and where computational and biological claims must be separated.

plain-language notes· study card
chapter 16 · Part III · Looking Deeper

Applications and Case Studies live

What landmark applications reveal about representation, simulation, search, engineering, evaluation, and the limits of algorithm-only explanations.

plain-language notes· study card
chapter 17 · Part III · Looking Deeper

Frontiers live

Open directions in prediction, temporal abstraction, state construction, reward design, continual learning, safety, and the societal effects of capable agents.

plain-language notes· study card