Entropy Regularization in Reinforcement Learning

In this post I assume you already know the basics of Reinforcement Learning (RL). If you are a beginner and would like to know more about RL, you can start with an introductory story of RL I wrote here.

In our everyday language, we commonly use the term “entropy” to refer to the lack of order or predictability of a system (for example, the universe.) In Reinforcement Learning (RL), the term is used in a similar fashion: in RL, entropy refers to the predictability of the actions of an agent. This is closely related to the certainty of its policy…