The Bellman Equation and Dynamic Programming
In the previous chapter, we learned that in reinforcement learning our goal is to find the optimal policy. The optimal policy is the policy that selects the correct action in each state so that the agent can get the maximum return and achieve its goal. In this chapter, we'll learn about two interesting classic reinforcement learning algorithms called the value and policy iteration methods, which we can use to find the optimal policy.
Before diving into the value and policy iteration methods directly, first, we will learn about the Bellman equation. The Bellman equation is ubiquitous in reinforcement learning and it is used for finding the optimal value and Q functions. We will understand what the Bellman equation is and how it finds the optimal value and Q functions.
After understanding the Bellman equation, we will learn about two interesting dynamic programming methods called value and policy iterations, which use...