The Past Present and Future of Artificial Neural Networks in Digital Games

作者: DK Charles , S McGlinchey

DOI:

关键词: Artificial neural networkComputer scienceVariety (cybernetics)Game DeveloperData scienceControl (management)Artificial perception

摘要: Over the past 20 years, topic of artificial neural networks has been a vibrant area AI research, leading to new algorithms that have used in variety disciplines including engineering, finance, perception and control & simulation. Despite this, there limited impact on commercial games industry. This paper reviews some successful uses identifies positive elements their use, discusses factors deterred use amongst game developers. Addressing these weaknesses, we outline ideas for future research may aid developers producing more convincing AI, supplement or replace traditional techniques.

参考文章(18)
Daniel Livingstone, Robert McDowell, Fast Marching and Fast Driving: Combining Off-Line Search and Reactive AI. GAME-ON. pp. 197- ,(2003)
Stephen J. McGlinchey, Learning of AI Players From Game Observation Data. GAME-ON. pp. 106- ,(2003)
Colin Fyfe, Dynamic Strategy Creation and Selection Using Artificial Immune Systems International Journal of Intelligent Games & Simulation. ,vol. 3, pp. 1- 6 ,(2004)
Quasim H. Mehdi, Norman E. Gough, Zhigang Wen, Neural Networks for Animating Variations in Character Behaviours. GAME-ON. ,(2002)
Ethan Rappaport, Arthur Swanberg, Interactive computer games ,(2001)
David Westbrook, Marc Atkin, Carole R. Beal, Paul Cohen, Joseph Beck, Intelligent modeling of the user in interactive entertainment ,(2002)
P. M. Sweetser, How to build neural networks for games Charles River Media. pp. 615- 625 ,(2004)
H. Yin, N.M. Allinson, Bayesian learning for self-organising maps Electronics Letters. ,vol. 33, pp. 304- 305 ,(1997) , 10.1049/EL:19970196