This is an extra-university project. I built a reinforcing learning model, that learns the complicated task of solving a minesweeper game. I tried different strategies (policy-based and DQN) and different network architectures. The final result is great. A video where the AI is playng: https://github.com/user-attachments/assets/4eb8ac15-77e6-4e4b-a8b1-43b638363ce4