alphazero-hive

This repository contains an implementation of the self-play based reinforcement learning algorithm for the game of Hive. It is based on DeepMind's paper on AlphaGo Zero, which can be found here. The self-play based engine builds upon the AlphaZero General framework. The engine uses Beekeeper for the game logic, an optimized implementation of Hive by Duncan Kampert. Both frameworks have been extended and modified to efficiently work together.