Games have always been a popular test bed for artificial intelligencetechniques. Game developers are always in constant search for techniques thatcan automatically create computer games minimizing the developer's task. Inthis work we present an evolutionary strategy based solution towards theautomatic generation of two player board games. To guide the evolutionaryprocess towards games, which are entertaining, we propose a set of metrics.These metrics are based upon different theories of entertainment in computergames. This work also compares the entertainment value of the evolved gameswith the existing popular board based games. Further to verify theentertainment value of the evolved games with the entertainment value of thehuman user a human user survey is conducted. In addition to the user survey wecheck the learnability of the evolved games using an artificial neural networkbased controller. The proposed metrics and the evolutionary process can beemployed for generating new and entertaining board games, provided an initialsearch space is given to the evolutionary algorithm.
展开▼
