Markov Decision Process and Reinforcement Learning

This chapter first provides the fundamental background and theory of the Markov decision process (MDP), a critical mathematical framework for modeling decision‐making in situations where outcomes are partially random and partially under the control of a decision‐maker. Specifically, key components of an MDP and several typical extension models are presented. After that, common solutions to address MDP problems such as linear programming, value iteration, policy iteration, and reinforcement learning, are reviewed.