Stephano Zanetti
ABSTRACT
This paper presents a First Person Shooter Artificial Intelligence system that makes use of machine learning capabilities to achieve more human-like behavior and strategies. The AI is trained with a supervised learning paradigm using example recorded during the observation of expert human players. The Machine Learning section of the AI is based on various Feed Forward Multi-layer Neural Networks trained by Genetic Algorithms. The AI system is developed and tested in the Quake 3 Arena game engine. The system is able to learn certain behaviors but still lack on some others. The results are evaluated and possible improvements are proposed.
- Benjamin G., 'An Empirical Study of Machine Learning Algorithms Applied to Modelling Player Behaviour in a First Person Shooter Video Game', University of Wisconsin -- Madison, USA. 2002.Google Scholar
- Johannes L., 'Einsatz von Neuronalen Netzen in einem Bot für Counterstrike' Friedrich Schiller-Gymnasium Preetz, 2001.Google Scholar
- Nicholas C., 'NeuralBot', 2003. http://homepages.paradise.net.nz/nickamy/neuralbot/index.shtmlGoogle Scholar
- Udo S., 'Multiple Layer Perceptron Training Using Genetic Algorithms', University of South Australia, 2001.Google Scholar
- Randall S. Sexton, Jatinder N. D. Gupta, 'Comparative evaluation of genetic algorithm and back-propagation for training neural networks', Southwest Missouri State University, USA, 2000.Google Scholar
- Mat B., Neural Network in Plain English, http://www.ai-junkie.com/nntl.html Accessed December 2002.Google Scholar
- Darrin B., Christopher G. Atkeson, 'Using Primitives in Learning From Observation', ATR-International, Information Sciences Division, Japan, College of Computing, Georgia Institute of Technology, USA, 2000.Google Scholar
- Paul T., 'First-Person Shooter Al Architecture', AI Game Programming Wisdom, Steve, R., Pg. 387--396, CHARLSE RIVER MEDIA INC., USA, 2002.Google Scholar
- J. M. P. van Waveren, 'The Quake III Arena Bot', University of Technology Delft, Faculty ITS, Holland, 2003.Google Scholar
- Russell S., Norvig P., Artificial Intelligence a modern approach, Prentice Hall, New Jersey, 2003. Google ScholarDigital Library
- Haykin S., Neural Networks, a comprehensive foundation, Prentice Hall, New Jersey, USA, 1994. Google ScholarDigital Library
- K. F. Man, K. S. Tang, S. Kwong, Genetic Algorithms, Springer, Great Britain, 1999.Google Scholar
- Talib S. H., 'Methods of Combining Neural Networks and Genetic Algorithms', Queen's University, Canada, 1997.Google Scholar
- World Cyber Games 2002, http://www.worldcybergames.com/, Korea, 2003.Google Scholar
- Daniel J., Janet W., 'Computer Games with Intelligence', 10th IEEE International Conference on Fuzzy system, Melbourne, Australia, 2001.Google Scholar
- Taha A. S, Avelino J. G., 'Learning situational awareness by observing expert actions', University of Central Florida, USA, 2000.Google Scholar
- Zanetti, S. 'Application of Machine Learning AI to First Person Shooter Games'. MSc Dissertation Computer Games Technology. Liverpool John Moores University, 2003.Google Scholar
Index Terms
- Machine learning techniques for FPS in Q3
Recommendations
Evolutionary design of constructive multilayer feedforward neural network
ICS'06: Proceedings of the 10th WSEAS international conference on SystemsThis paper proposes an evolutionary design methodology of multilayer feedforward neural networks based on constructive approach. We elaborate an adjustable processing element as primitive neuron model. The neural layer can be constructed by assembling ...
Cerberus: applying supervised and reinforcement learning techniques to capture the flag games
AIIDE'08: Proceedings of the Fourth AAAI Conference on Artificial Intelligence and Interactive Digital EntertainmentApplying machine learning techniques to real-time computer games is an active research field. In this paper we present Cerberus, a machine learning framework for team-based Capture The Flag (CTF) games. This framework utilizes reinforcement learning to ...
On removing conflicts for machine learning
AbstractA Machine Learning (ML) system learns from a set of samples called the training set. In some cases, the training set may have learning conflicts that affect the performance of the machine learning system. A learning conflict is ...
Highlights- Learning conflicts that affect machine learning system performance.
- Samples in ...
Comments