Reinforcement Learning Algorithms with Python

Summary

Hopefully, in this chapter, you have learned about all the tools and components needed to build RL algorithms. You set up the Python environment required to develop RL algorithms and programmed your first algorithm using an OpenAI Gym environment. As the majority of state-of-the-art RL algorithms involve deep learning, you have been introduced to TensorFlow, a deep learning framework that you'll use throughout the book. The use of TensorFlow speeds up the development of deep RL algorithms as it deals with complex parts of deep neural networks such as backpropagation. Furthermore, TensorFlow is provided with TensorBoard, a visualization tool that is used to monitor and help the algorithm debugging process.

Because we'll be using many ...