In this chapter, you were introduced to the basic concepts of reinforcement learning. After getting acquainted with the OpenAI Gym toolkit, you were presented with the MountainCar challenge, where a car needs to be controlled in a way that will allow it to climb the taller of two mountains. After solving this challenge using genetic algorithms, you were introduced to the next challenge, CartPole, where a cart is to be precisely controlled to keep an upright pole balanced. We were able to solve this challenge by combining the power of a neural network-based controller with genetic algorithm-guided training.
In the next chapter, we will transition to the world of art and find out how genetic algorithms can be used to reconstruct images of famous paintings with a set of semi-transparent overlapping shapes.