Centurio, a General Game Player: Parallel, Java- and ASP-based

Abstract We present the General Game Playing system Centurio. Centurio is a Java-based player featuring different strategies based on Monte Carlo Tree Search extended by techniques borrowed from Upper Confidence bounds applied to Trees as well as Answer Set Programming (for single-player games). Centurio’s Monte Carlo Tree Search is accomplished in a massively parallel way by means of multi-threading as well as clustercomputing. Another major feature of Centurio is its compilation of game descriptions, states, and state manipulations into Java, yielding an edge over existing Prolog-based approaches. Centurio is open source software freely available via the web. Keywords Answer set programming · General game playing · Monte Carlo tree search · Parallelization

• Type: Journal
• Year: 2011
• Where: KI
• Authors: Maximilian Möller, Marius Thomas Schneider, Martin Wegner, Torsten Schaub
• Keywords: Artificial Intelligence, Carlo Tree Search, General Game Playing, KI 2011, Monte Carlo Tree