Summary

In this chapter, we started our journey into the RL world by learning what makes RL special and how it relates to the supervised and unsupervised learning paradigm. We then learned about the basic RL formalisms and how they interact with each other, after which we defined Markov process, Markov reward process, and Markov decision process.

In the next chapter, we'll move away from the formal theory into the practice of RL. We'll cover the setup required, libraries, and write our first agent.