Learning to play Pac-Man: An evolutionary, rule-based approach

Gallagher, M. R. and Ryan, A. J. (2003). Learning to play Pac-Man: An evolutionary, rule-based approach. In: R. Sarker, R. Reynolds and H. Abbass, 2003 Congress on Evolutionary Computation. The 2003 Congress on Evolutionary Computation (CEC 2003), Canberra, Australia, (2462-2469). 8-12 December 2003. doi:10.1109/CEC.2003.1299397


Author Gallagher, M. R.
Ryan, A. J.
Title of paper Learning to play Pac-Man: An evolutionary, rule-based approach
Conference name The 2003 Congress on Evolutionary Computation (CEC 2003)
Conference location Canberra, Australia
Conference dates 8-12 December 2003
Proceedings title 2003 Congress on Evolutionary Computation
Journal name Cec: 2003 Congress On Evolutionary Computation, Vols 1-4, Proceedings
Place of Publication Piscataway, NJ, U.S.A.
Publisher The Institute of Electrical and Electronics Engineers
Publication Year 2003
Sub-type Fully published paper
DOI 10.1109/CEC.2003.1299397
ISBN 0-7803-7805-9
Editor R. Sarker
R. Reynolds
H. Abbass
Volume 4
Start page 2462
End page 2469
Total pages 8
Collection year 2003
Language eng
Abstract/Summary Pac-Man is a well-known, real-time computer game that provides an interesting platform for research. We describe an initial approach to developing an artificial agent that replaces the human to play a simplified version of Pac-Man. The agent is specified as a simple finite state machine and ruleset. with parameters that control the probability of movement by the agent given the constraints of the maze at some instant of time. In contrast to previous approaches, the agent represents a dynamic strategy for playing Pac-Man, rather than a pre-programmed maze-solving method. The agent adaptively "learns" through the application of population-based incremental learning (PBIL) to adjust the agents' parameters. Experimental results are presented that give insight into some of the complexities of the game, as well as highlighting the limitations and difficulties of the representation of the agent.
Subjects E1
280212 Neural Networks, Genetic Alogrithms and Fuzzy Logic
780101 Mathematical sciences
Keyword Computer games
Evolutionary computation
Finite state machines
Knowledge based systems
Q-Index Code E1

 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 15 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 34 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Fri, 24 Aug 2007, 01:57:59 EST