We propose the popular board game Blood Bowl as a new challenge for Artificial Intelligence (AI). Blood Bowl is a fully-observable, stochastic, turn-based, modern-style board game with a grid-based playing board. At first sight, the game ought to be approachable by numerous game-playing algorithms. However, as all pieces on the board belonging to a player can be moved several times each turn, the turn-wise branching factor becomes overwhelming for traditional algorithms. Additionally, scoring points in the game is rare and difficult, which makes it hard to design heuristics for search algorithms or apply reinforcement learning. We present our work in progress on a game engine that implements the core rules of Blood Bowl with a forward model and a reinforcement learning interface. We plan to release the engine as open source and use it to facilitate future AI competitions.
|Blood Bowl: The Next Board Game Challenge for AI
|Association for Computing Machinery
|Udgivet - 2018
|FDG Workshop on Tabletop Games - Malmö, Sverige
Varighed: 7 aug. 2018 → 10 aug. 2018
|FDG Workshop on Tabletop Games
|07/08/2018 → 10/08/2018