Uniform versus Gaussian mutators in automatic generation of game AI in Ms. Pac-man using hill-climbing

Author

Tan, Tse Guan ; Teo, Jason ; Anthony, Patricia

Author_Institution

Evolutionary Comput. Lab., Univ. Malaysia Sabah, Kota Kinabalu, Malaysia

fYear

2010

fDate

17-18 March 2010

Firstpage

282

Lastpage

286

Abstract

This paper explores the idea of combining the hill-climbing concept into feed-forward artificial neural networks (ANN) to develop intelligent controllers to play the Ms. Pacman game. The resulting algorithm is referred to as the HillClimbingNet. A comparison with a random system, called RandNet is conducted on the same problem. We also present a survey of the effects of two most popular probability density functions, uniform and Gaussian distributions/mutators on the introduced algorithm. The results clearly indicate the strong potential of the hill-climbing strategy as a direct search method in tandem with a Gaussian-based mutator to optimize the ANN for playing Ms. Pac-Man.

Keywords

Gaussian distribution; computer games; feedforward neural nets; AI game; Gaussian mutator; HillClimbingNet algorithm; Ms. Pacman game; RandNet random system; feedforward artificial neural networks; hill-climbing concept; probability density functions; uniform mutator; Artificial intelligence; Artificial neural networks; Automatic control; Feedforward systems; Gaussian distribution; Gaussian processes; Intelligent networks; Optimization methods; Probability density function; Search methods; Gaussian-based mutator; Ms. Pac-man; feed-forward artificial neural networks; hill-climbing;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Retrieval & Knowledge Management, (CAMP), 2010 International Conference on

Conference_Location

Shah Alam, Selangor

Print_ISBN

978-1-4244-5650-5

Type

conf

DOI

10.1109/INFRKM.2010.5466903

Filename

5466903