An AI for quoridor based on Alpha Zero. Please check out the architecture behind the algorithm here.
A concise description of our algorithm can be found here.
While the current code is fairly functional, we could benefit from the following contributions:
- Asynchronous MCTS as described in the paper
- An asynchronous version of the code- parallel processes for self-play, neural net training and model comparison.
- Game logic files for more games that follow the specifications in
Game.py, along with their neural networks - Neural networks in other frameworks
- Pre-trained models for different game configurations
- Leonardo_Ramos Author of this implementation of quoridor AI.
- Shantanu Thakoor and Megha Jhunjhunwala helped with core design and implementation.