Multi-scale Bayesian modeling for RTS games: an application to StarCraft AI

Gabriel Synnaeve (1) , Pierre Bessiere (2)

1 LSCP - Laboratoire de sciences cognitives et psycholinguistique
2 LPPA - Laboratoire de Physiologie de la Perception et de l'Action

This paper showcases the use of Bayesian models for real-time strategy (RTS) games AI in three distinct core components: micro-management (units control), tactics (army moves and positions), and strategy (economy, technology, production, army types). The strength of having end-to-end probabilistic models is that distributions on specific variables can be used to inter-connect different models at different levels of abstraction. We applied this modeling to StarCraft, and evaluated each model independently. Along the way, we produced and released a comprehensive dataset for RTS machine learning.

Mots clés: Bayesian modeling; RTS AI; real-time strategy; video games; StarCraft; tactics; micro-management

Domaines: Informatique; Intelligence artificielle [cs.AI]; Informatique et théorie des jeux [cs.GT]

Gabriel Synnaeve, Pierre Bessiere. Multi-scale Bayesian modeling for RTS games: an application to StarCraft AI. IEEE Transactions on Computational Intelligence and AI in games, 2015, ⟨10.1109/TCIAIG.2015.2487743⟩. ⟨hal-01228961⟩