Experimental Results on Alternative Strategies in Two Agent-Based Game Playing Systems

12:00noon, Sep 25, Thursday, 2008, ST2, 430


Andrés Gómez de Silva Garza
Computer Engineering Department
Mexican Autonomous Technology Institute (ITAM)


In this talk I will discuss two games and the agent-based systems I have implemented to play them. The nature of the two games is quite different. The first one is played by teams, and teammates cannot communicate with each other during game play, yet must collaborate in order to compete successfully against the opposing team. The second one is a board game in which most steps require deciding from among several alternative moves. I describe several alternate strategies for making decisions on what moves to make that have been included in both implementations. I will then present a set of simple experiments designed to evaluate the alternate strategies.

Short Bio

Andrés Gómez de Silva Garza has an undergraduate degree in Computer Engineering from the National Autonomous University of Mexico (UNAM), a Master's degree in Computer Science from the Georgia Institute of Technology, and a Ph.D. from the University of Sydney in Australia. Since 2000 he has been teaching and doing research in the Computer Engineering Department of the Mexican Autonomous Technology Institute (ITAM). His research background is on case-based reasoning and evolutionary algorithms applied to design and creativity, and agent-based game playing systems with multiple strategies.