Reinforcement learning (RL) has gained considerable traction recently. It is a different approach to machine intelligence from traditional machine learning and deep learning techniques. It has achieved human-level performance in learning complex games such as Go and Dota. RL is an artificial intelligence framework where an agent performs learning through trial and error. It is a learning process that mimics the fundamental way humans learn. The overarching goal of this chapter is to make you conversant with the components of RL. You will learn how to implement RL using various packages in R.
In this chapter, we will cover the following recipes:
- Model-based RL using MDPtoolbox
- Model-free RL
- Cliff walking using RL
