This repo provides an implementation of self-play based reinforcement learning based on the AlphaZero paper.
Please note: This repo is work in progress. The basic structure of this repo is inspired by this repo. However, the code is adapted such that it can handle the more general setup of Scotland Yard (i.e. N-player, asymmetric game)