A simple heuristic search method for the automatic generation of neural-based game artificial intelligence architectures in Ms. Pac-Man

Author

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

Author_Institution

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

fYear

2010

fDate

10-13 May 2010

Firstpage

753

Lastpage

756

Abstract

In this work, we develop a game controller called HillClimbingNet (Hill-Climbing Neural Network) for playing Ms. Pac-man that combines the hill-climbing concept and simple feedforward neural network. The computational experiments have been conducted to evaluate and compare the proposed algorithm against Random Direction (RandDir) and Random Neural Network (RandNet) systems. According to the simulation results, HillClimbingNet has achieved an average score of 6290, but only 439 and 735 on the RandDir and RandNet, respectively. HillClimbingNet has a very good performance.

Keywords

computer games; evolutionary computation; feedforward neural nets; heuristic programming; neural nets; stochastic games; unsupervised learning; Ms. Pac-man; RandDir; RandNet; artificial intelligence architecture; automatic generation; feedforward neural network; game controller; heuristic search method; hill-climbing neural network; neural-based game; random direction; random neural network system; Artificial neural networks; Feedforward neural networks; Games; Intelligent systems; Search methods; Unsupervised learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Sciences Signal Processing and their Applications (ISSPA), 2010 10th International Conference on

Conference_Location

Kuala Lumpur

Print_ISBN

978-1-4244-7165-2

Type

conf

DOI

10.1109/ISSPA.2010.5605407

Filename

5605407