In this chapter, we took a deep dive into the inner workings of more sophisticated RL algorithms, such as DQN and PPO. We started by walking through the installation of the Python tools and dependencies, where we learned how to use the more basic tools, such as Jupyter Notebook. Then we built a working ML-Agents example that used an external Python agent's brain. After that, we covered the basics of neurons and neural networks. From there, we took a look at DQN and a basic deep Q-learning agent using Keras. We completed the chapter by looking at another RL algorithm called PPO. As we learned, PPO will be the workhorse for many of our complex situations.
Our journey this chapter was more or less a setup for the next chapter, where we start to dig in deep and build on the foundations we laid in this chapter. We will take a closer look at PPO and how it can drive other...
