Skip to main content

Ctrl+K

Contents

I: INTRODUCTION

Foreword
What is reinforcement learning?
Getting started with a first example

II: VALUE-BASED METHODS

Introduction to value-based methods
Markov Decision Processes
Value Iteration
Multi-armed bandits
Temporal difference reinforcement learning
Monte-Carlo Tree Search (MCTS)

III: POLICY-BASED METHODS

Introduction to policy-based methods
Policy iteration
Policy gradients
Actor-critic methods

IV: SCALING

Q-function approximation
Reward shaping
n-step reinforcement learning
Experience replay
Modelling and abstraction for MDPs

V: MULTI-AGENT REINFORCEMENT LEARNING

Normal form games
Extensive form games
Backward induction
Multi-agent reinforcement learning
Modelling and abstraction for multi-agent games

VI: APPENDIX

Introduction to basic probability theory

Repository
Open issue

Index

By Tim Miller, The University of Queensland

© Copyright 2023.