Summary

In this chapter, we have successfully built a deep reinforcement learning model, each with Q-learning and SARSA learning in Keras using the CartPole game from OpenAI Gym. We understood Q-learning, SARSA learning, how to interact with game environments from Gym, and the function of the agent (deep learning model). We defined some key hyperparameters, as well as, in some places, reasoned with why we used what we did. Finally, we tested the performance of our reinforcement learning on new games and determined that we succeeded in achieving our goals.