Advanced Exploration
In this chapter, we will talk about the topic of exploration in reinforcement learning (RL). It has been mentioned several times in the book that the exploration/exploitation dilemma is a fundamental thing in RL and very important for efficient learning. However, in the previous examples, we used quite a trivial approach to exploring the environment, which was, in most cases, π-greedy action selection. Now itβs time to go deeper into the exploration subfield of RL, as more complicated environments might require much better exploration strategies than π-greedy approach.
More specifically, we will cover the following key topics:
-
Why exploration is such a fundamental topic in RL
-
The effectiveness of the epsilon-greedy (π-greedy) approach
-
Alternatives and how they work in different environments
We will implement the methods described to solve...
