Reinforcement Learning
Markov Decision Processes, Q-Learning, Policy Gradients and Deep RL
About this guide
A concise exam companion for reinforcement learning. It builds from Markov decision processes, returns and value functions through the Bellman equations, dynamic programming, Monte Carlo and temporal-difference learning, Q-learning and SARSA, to function approximation, policy gradients and actor-critic methods, deep Q-networks and learning from human feedback. Every chapter ends with a worked example carried through step by step, so you can check the reasoning yourself.
Inside the guide
- Markov decision processes
- Returns and value functions through the Bellman equations
- Dynamic programming
- Monte Carlo and temporal-difference learning
- Q-learning and SARSA
- To function approximation
- Policy gradients and actor-critic methods
- Deep Q-networks and learning from human feedback
Questions students ask
What does Reinforcement Learning cover?
Reinforcement Learning covers the core exam topics.
Who is Reinforcement Learning for?
It is written for university students revising reinforcement learning, and for lecturers and librarians choosing a clear, exam-focused reading-list text.
What format does it come in?
Kindle ebook and paperback on Amazon.
Can I read part of it for free?
Full Marks Press offers a free sample chapter at fullmarkspress.com/free, and the full guide is free to read with Kindle Unlimited.