ConnectZero: Understanding Self-Play in Connect 4

This project explores the dynamics of self-play in the Connect 4 environment using reinforcement learning agents. The goal is to compare different self-play settings and analyze the learning process and agent performance.

This project is part of the INF8250AE -- Reinforcement Learning course at Polytechnique Montréal in 2024:

• Edouard Albert-Roulhac
• David Heurtel-Depeiges
• Arthur Toulouse

Environment Setup

1. Clone the repository:

   git clone https://github.com/edarsem/ConnectZero.git
   cd ConnectZero

2. Create and activate the virtual environment:

    python3 -m venv ConnectZero_env
    source ConnectZero_env/bin/activate

3. Install dependencies:

    pip install -r requirements.txt -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html