This repository is a simple implementation of the Monte Carlo Tree Search Algorithm using Atari Environments. This implementation currently uses the basic Upper Confidence Bound (UCB1) formula.
To run the code you will need to install these packages:
conda create -n MCTS python=3.10
conda activate MCTS
pip install gymnasium[atari]
pip install gymnasium[accept-rom-license]
If you would like to run the python implementation run:
python MonteCarloTreeSearch.py
There is also a notebook MonteCarloTreeSearch.ipynb
- Detailed explanation of Monte Carlo Tree Search
- Potentially update with more advanced methods
- Create visualization for tree