Adjutant bot: An evaluation of unit micromanagement tactics

Constructing an effective real-time strategy bot requires multiple interlocking elements including a well-designed architecture, efficient build order, and good strategic and tactical decision-making. However even when the bot's high-level strategy and resource allocation is sound, poor battlefield tactics can result in unnecessary losses. This paper focuses on the problem of avoiding troop loss by identifying good tactical groupings. Banding separated units together using UCT (Upper Confidence bounds applied to Trees) along with a learned reward model outperforms grouping heuristics at winning battles while preserving resources. This paper describes our findings in the context of the Adjutant bot design which won the best Newcomer honor at CIG 2012 and is the basis for our 2013 entry.