Bayesian Reinforcement Learning

Discusses models and methods for Bayesian inference in the simple single-step Bandit model. The book then reviews the extensive recent literature on Bayesian methods for model-based RL, where prior information can be expressed on the parameters of the Markov model.