Combat Models for RTS Games
This work addresses a specific bottleneck for AI researchers and developers in RTS games, but it is incremental as it builds on existing forward model concepts.
The paper tackled the problem of lacking forward models for game tree search in real-time strategy (RTS) games by developing three combat models for attrition games, showing they can be learned from replay data and used for tactical decisions in StarCraft, with experiments reporting their predictive accuracy and impact.
Game tree search algorithms, such as Monte Carlo Tree Search (MCTS), require access to a forward model (or "simulator") of the game at hand. However, in some games such forward model is not readily available. This paper presents three forward models for two-player attrition games, which we call "combat models", and show how they can be used to simulate combat in RTS games. We also show how these combat models can be learned from replay data. We use StarCraft as our application domain. We report experiments comparing our combat models predicting a combat output and their impact when used for tactical decisions during a real game.