AlphaZero Duck Chess

Using reinforcement learning to train an AI for the game of Duck Chess. The model learns solely through self-play, using an algorithm based on AlphaZero.