Entry year: 2005
Degree: BCompSci
Enrolment status: checked and confirmed
Degree status: complete semester 1, 2006
Mark Wittkamp
wittkm01@csse.uwa.edu.au
Project: Application of Evolutionary Computation to Opponent Modeling in Games of Imperfect Information
Supervisor(s): Luigi Barone
Project status: complete semester 1, 2006
Games of imperfect information are games where not all information concerning the state of play are known (for example, hidden opponent cards in Poker). Due to the non-deterministic nature of these games, satisfactory solutions for playing them are often very difficult to construct. Evolutionary Algorithms (EAs) seem to lend themselves to these types of problems because of their ability to adapt to a changing environment. In the past, EAs have been witness to success for opponent modeling in games of imperfect information. This research is to be conducted upon the game of Spoof: a coin-game played by 2 or more players. This game has a lot of potential for strategy development ranging from simple probability to opponent inference, bluffing and adaptive behaviour. The aim of this research is to explore various existing and new instances of EAs to create an artificial opponent that is able to adapt and respond to various game-playing strategies for the game of Spoof. A further goal is to make observations as to the suitability of these techniques in other games.
Last update: Thu Aug 3 00:39:11 2006