02062006

Title:  Evolving Adaptive Play for the Game of Spoof

Date:    11am, Friday 2 June 2006
Speaker: Mark Wittkamp

In games where no optimal fixed strategy is known to exist, it is important for players to be adaptive.  Adaptive artificial opponents capable of learning and opponent modelling are highly desirable in computer games.  Typically, a great deal of a game's ability to maintain the interest of human players is provided by multiplayer functionality due to the unpredictable and changing game environment that this entails. It is reasonable to expect that artificial opponents mimicking the observable characteristics of human players through adaptive play would significantly benefit many games' lastability.  Spoof is a multiple player game of imperfect information for which the success of a player is largely dictated by its ability to build models of its opponent(s) so that their weaknesses may be identified and exploited.

We present our approach to opponent modelling in the game of Spoof using Evolutionary Computation, more specifically - Genetic Programming.Genetic Programming involves a guided random search of the solution space to a given problem by evolving a population of candidate solutions which take the form of program trees.  Genetic Programming
shows potential for games of imperfect information or other games where tree searching algorithms are often infeasible due to the games' intractability.
The suitability of Genetic Programming for opponent modelling is substantiated by comparison with a simple lookup-table approach for learning.  We demonstrate that specialisation and opponent modelling is required for optimal play in the game of Spoof by contrasting
evolved playing strategies with a number of fixed strategies comparable to those employed by most human players.

Title:  A Coupled Atmosphere-Fire Model Using Cellular Automata

Date:    11:30am, Friday 2 June 2006
Speaker: Richard Hart

Because of their complex nature, bushfires are very difficult to predict and fight.  There are many different types of bushfire simulation models, from theoretical, which are computationally complex, to empirical, which are less accurate.  Some theoretical models consider the interaction between the bushfire and the atmosphere, which are more accurate than those models which consider the fire alone.  A different approach is to use cellular automata to model a ground fire, which discretises areas of the fire into finite states to reduce the computational complexity of theoretical models.

The goal of this project is to develop an experimental tool to investigate the viability of combining these two approaches, using cellular automata to implement a coupled atmosphere-fire model in a bushfire simulator.The focus of the development is to produce software which is simple to modify, as it is to be used by researchers to develop and improve the model.