Solving the Markov Decision Process

Reinforcement learning is one of the most exciting fields in machine learning. There are many machine learning systems leveraged by reinforcement learning algorithms. AlphaGo is the famous example for this type of application. So far, we have seen applications using supervised learning and unsupervised learning techniques. While they are powerful in specific fields, they do not provide a distinct solution where even a human does not know the correct answer. In this chapter, we are going to introduce reinforcement learning, which has become one of the most prominent fields recently. This technology has shown outstanding performance when there is no complete knowledge about the target, such as in TV games. We will also learn how to implement the Bellman equation for solving Markov decision process (MDP) problems and how it relates to reinforcement...