Biologically Inspired Artificial Intelligence for Computer Games

by Charles, Darryl

• ISBN13: 9781591406464
• ISBN10: 1591406463
• Format: Hardcover
• Copyright: 2007-11-30
• Publisher: Igi Global

Summary

Computer games are often played by a human player against an artificial intelligence software entity. In order to truly respond in a human-like manner, the artificial intelligence in games must be adaptive, or respond as a human player would as he/she learns to play a game.Biologically Inspired Artificial Intelligence for Computer Games reviews several strands of modern artificial intelligence, including supervised and unsupervised artificial neural networks; evolutionary algorithms; artificial immune systems, swarms, and shows-using case studies for each to display how they may be applied to computer games. This book spans the divide which currently exists between the academic research community working with advanced artificial intelligence techniques and the games programming community which must create and release new, robust, and interesting games on strict deadlines, thereby creating an invaluable collection supporting both technological research and the gaming industry.

Table of Contents

Foreword	p. vii
Preface	p. ix
Acknowledgment	p. xiv
Contemporary Video Game AI	p. 1
Introduction	p. 1
The Dawn of the Computer Video Game	p. 1
Contemporary Video Game AI	p. 5
Conclusion	p. 10
References	p. 10
An Introduction to Artificial Neural Networks	p. 12
Introduction	p. 12
Biological Neural Networks	p. 13
Artificial Neural Networks	p. 14
Neural Networks Classification	p. 17
Learning in Artificial Neural Networks	p. 21
References	p. 23
Supervised Learning with Artificial Neural Networks	p. 24
Introduction	p. 24
The Delta Rule	p. 24
Multipayered Perceptrons	p. 28
The Backpropagation	p. 28
Issues in Backpropagation	p. 31
An Example	p. 36
References	p. 39
Case Study: Supervised Neural Networks in Digital Games	p. 41
Introduction	p. 41
Robocode	p. 41
Conclusion	p. 46
References	p. 46
Unsupervised Learning in Artificial Neural Networks	p. 48
Unsupervised Learning	p. 48
Hebbian Learning	p. 49
Hebbian Learning and Information Theory	p. 51
Anti-Hebbian Learning	p. 59
Independent Component Analysis	p. 61
A Case Study: Independent Component Analysis	p. 62
Competitive Learning	p. 68
Applications	p. 77
Case Study: The Self-Organizing Map and Pong	p. 77
Conclusion	p. 89
References	p. 89
Fast Learning in Neural Networks	p. 91
Introduction	p. 91
Radial Basis Functions	p. 91
Error Descent	p. 100
Pong: A Comparative Study, MLP vs. RBF	p. 101
Conclusion	p. 104
Reference	p. 104
Endnote	p. 104
Genetic Algorithms	p. 105
Introduction	p. 105
Genetic Algorithms	p. 106
A First Example	p. 114
Case Study: GA and Backproagation ANN for Motocross Controllers	p. 116
Futher Reading	p. 118
Summary	p. 119
References	p. 120
Beyond the GA: Extensions and Alternatives	p. 121
Introduction	p. 121
The Iterated Prisoners' Dilemma (IPD)	p. 130
N-Persons Iterated Prisoners' Dilemma (NIPD)	p. 132
References	p. 138
Evolving Solutions for Multiobjective Problems and Hierarchical AI	p. 139
Introduction	p. 139
Multiobjective Problems	p. 140
Coevolution in Hierarchical AI for Strategy Games	p. 141
Conclusion	p. 147
References	p. 148
Artificial Immune Systems	p. 150
Introduction	p. 150
The Immue System	p. 151
Artificial Immune Systems	p. 153
Hypermutations	p. 156
The Immune Network	p. 158
Agent Wars	p. 160
Cooperating Strategies	p. 162
Incomplete Information	p. 164
Changing the Game	p. 166
Duelling	p. 167
Discussion	p. 167
Comparision with Gas	p. 170
Conclusion	p. 178
References	p. 178
Ant Colony Optimisation	p. 180
Introduction	p. 180
Foraging Behaviour of Ants	p. 181
Simulating Artificial Ant Colonies with the S-ACO Algorithm	p. 183
Improvements to the S-ACO Algorithm	p. 188
Case Study: S-ACO and Combat	p. 193
Summary	p. 199
Further Reading	p. 200
References	p. 200
Reinforcement Learning	p. 202
Introduction	p. 202
The Main Elements	p. 203
Finding the Best Policy	p. 205
Temporal Difference Learning	p. 211
TD([lambda] Methods	p. 213
Continuous State Spaces	p. 214
Immediate Rewards	p. 218
The Bernoulli Learner	p. 219
The Gaussian Learner	p. 221
Application to Games	p. 223
Comparison with Other Methods	p. 224
Conclusion	p. 225
Further Reading	p. 225
References	p. 226
Adaptivity within Games	p. 227
Introduction	p. 227
Existing Approaches to Adaptivity in Games	p. 229
Adaptive Technologies	p. 229
Developing Adaptive Solutions	p. 234
Emergent Gameplay	p. 236
Conclusion	p. 236
References	p. 237
Turing's Test and Believable AI	p. 239
Introduction: Contemporary Game AI	p. 239
The Popular Turing Test	p. 240
The Real Turing Test?	p. 242
Playing the Turing Test	p. 243
Generalising the Turing Test	p. 243
Measuring Believability	p. 245
Believable Computer Players	p. 247
Unbelievable Characters	p. 251
The Embedded Turing Test	p. 251
Conclusion	p. 253
Further Reading	p. 254
References	p. 254
About the Contributors	p. 256
Index	p. 258
Table of Contents provided by Ingram. All Rights Reserved.