This chapter presents the concept of reinforcement learning, which is widely used in gaming and robotics. The second part of this chapter is dedicated to learning classifier systems, which combine reinforcement learning techniques with evolutionary computing introduced in the previous chapter. Learning classifiers are an interesting breed of algorithms that are not commonly included in literature dedicated to machine learning. I highly recommend that you to read the seminal book on reinforcement learning by R. Sutton and A. Barto [11:1] if you are interested to know about the origin, purpose, and scientific foundation of reinforcement learning.
In this chapter, you will learn the following topics:
Basic concepts behind reinforcement learning
A detailed implementation of the Q-learning algorithm
A simple approach to manage and balance an investment portfolio using reinforcement learning
An introduction to learning classifier systems
A simple implementation of...