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 environments in the subsequent chapters, it's important to have a clear understanding of their differences and distinctiveness. By now, you should also be able to choose the best environments for your own projects, but bear in mind that despite the fact that we provided you with a comprehensive list, there may be many others that could better suit your problem.

That being said, in the following chapters, you'll finally learn how to develop RL algorithms. Specifically, in the next chapter, you will be presented with algorithms that can be used in simple problems where the environment is completely known. After those, we'll build more sophisticated ones that can deal with more complex cases.