📄️ Introduction
This is the Kaggle competition regarding Game AI and Reinforcement Learning.
📄️ Environment
I made the class for the environment using OpenAI gymnasium and initialized an environment.
📄️ Self-play
The policy is a custom CNN policy made using PyTorch:
📄️ Agent
The final policy is used to make the agent, which like the analytical agent, follows some rules.