This project implements a reinforcement learning model using the Advantage Actor-Critic (A2C) algorithm to solve the Atari AirRaid environment. The model utilizes a Convolutional Neural Network (CNN) policy for learning and playing the game.
Before running the code, ensure you have the following dependencies installed:
- Python 3.x
- Gymnasium
- Pytorch
- Stable Baselines3
Install these dependencies The training logs are present in the repository.