Reinforcement Learning with Genetic Algorithms
In this chapter, we will demonstrate how genetic algorithms can be applied to reinforcement learning – a fast-developing branch of machine learning that is capable of tackling complex tasks. We will do this by solving two benchmark environments from the Gymnasium (formerly OpenAI Gym) toolkit. We will start by providing an overview of reinforcement learning, followed by a brief introduction to Gymnasium, a toolkit that can be used to compare and develop reinforcement learning algorithms, as well as a description of its Python-based interface. Then, we will explore two Gymnasium environments, MountainCar and CartPole, and develop genetic algorithm-based programs to solve the challenges they present.
In this chapter, we will cover the following topics:
- Understanding the basic concepts of reinforcement learning
- Becoming familiar with the Gymnasium project and its shared interface
- Using genetic algorithms to solve...
