In this project, we covered the concepts of reinforcement learning and why they're popular among developers for creating game-playing AIs. We discussed AlphaGo and its sibling projects by Google DeepMind and studied their working algorithms in depth. Next, we created a similar program for playing Connect 4 and then for chess. We deployed the AI-powered chess engine to GCP on a GPU instance as an API and integrated it with a Flutter-based app. We also learned about how UCI is used to facilitate stateless gameplay for chess. After this project, you are expected to have a good understanding of how we can convert games into reinforcement learning environments, how to define gameplay rules programmatically, and how to create self-learning agents for playing these games.
In the next chapter, we will create an app that can make low-resolution images very high-resolution...