Introduction to Reinforcement Learning

Foundations and Trends monograph covering DQN, policy gradient, actor-critic methods, and practical challenges in deep RL from McGill and Google Brain researchers.

Vincent François-Lavet, Peter Henderson, Riashat Islam, Marc G. Bellemare, Joelle Pineau

Concise mathematical treatment of RL algorithms covering TD learning, policy gradient, and value function approximation from a leading RL theorist.

Csaba Szepesvári