Summary

In this chapter, we learned about RL systems. We discussed the premise of RL and how we can set it up. We talked about the differences between RL and supervised learning. We went through some real-world examples of RL and saw how various systems use it in different forms.

We discussed the building blocks of RL and concepts such as agent, environment, policy, reward, and so on. We then created an environment in Python to see it all in action. Finally, we used these concepts to build an RL agent.

In the next chapter, we will go into a rather different topic and learn how big data technologies can help us make our machine learning systems more robust and efficient.