Multiobjective Exploration of the StarCraft Map Space

Document type: Conference Papers
Peer reviewed: Yes
Full text:
Author(s): Julian Togelius, Mike Preuss, Nicola Beume, Simon Wessing, Johan Hagelbäck, Georgios N. Yannakakis
Title: Multiobjective Exploration of the StarCraft Map Space
Conference name: 2010 IEEE Conference on Computational Intelligence and Games (CIG)
Year: 2010
Pagination: 265
ISBN: 978-1-4244-6295-7
Publisher: IEEE
City: Copenhagen
URI/DOI: 10.1109/ITW.2010.5593346
Organization: Blekinge Institute of Technology
Department: School of Computing (Sektionen för datavetenskap och kommunikation)
School of Computing S-371 79 Karlskrona
+46 455 38 50 00
http://www.bth.se/com
Authors e-mail: julian@togelius.com, mike.preuss@cs.tu-dortmund.de, nicola.beume@cs.tu-dortmund.de, simon.wessing@cs.tu-dortmund.de, johan.hagelback@bth.se, yannakakis@itu.dk
Language: English
Abstract: This paper presents a search-based method for generating maps for the popular real-time strategy (RTS) game StarCraft. We devise a representation of StarCraft maps suitable for evolutionary search, along with a set of fitness functions based on predicted entertainment value of those maps, as derived from theories of player experience. A multiobjective evolutionary algorithm is then used to evolve complete StarCraft maps based on the representation and selected fitness functions. The output of this algorithm is a Pareto front approximation visualizing the tradeoff between the several fitness functions used, and where each point on the front represents a viable map. We argue that this method is useful for both automatic and machine-assisted map generation, and in particular that the Pareto fronts are excellent design support tools for human map designers
Subject: Computer Science\Artificial Intelligence
Digital Game Development\General
Keywords: Real-time strategy games, RTS, procedural content generation, evolutionary multiobjective optimization
Edit