Chapter 12. Reinforcement learning

We have covered supervised and unsupervised learning methods in-depth in Chapter 5, Decision Tree based learning, with various algorithms. In this chapter, we will be covering a new learning technique that is different from both supervised and unsupervised learning called Reinforcement Learning (RL). Reinforcement Learning is a particular type of Machine learning where the learning is driven by the feedback from the environment, and the learning technique is iterative and adaptive. RL is believed to be closer to human learning. The primary goal of RL is decision making and at the heart of it lies Markov's Decision Process (MDP). In this chapter, we will cover some basic Reinforcement Learning methods like Temporal Difference (TD), certainty equivalence, policy gradient, dynamic programming, and more. The following figure depicts different data architecture paradigms that will be covered in this chapter:

The following topics are covered...