원문정보
초록
영어
Computer game is a vibrant research area in artificial intelligence. Chinese chess game is an important part of computer game and it has become an important study area after chess game had reached its culmination when Deep Blue and its successors beat Kasparov. Some achievements acquired in Chinese chess game have applied into fields of medicine, economics and military. This paper presented a new method of optimizing evaluation function in Chinese-chess programming by particle swarm optimization. The process of training evaluation function is to automatically adjust these parameters in the evaluation function by self-optimizing method accomplished through competition, which is a Chinese-chess system plays against itself with different evaluation functions. The results show that the particle swarm optimization is successfully applied to optimize the evaluation function in Chinese chess and the performance of the presented program is effectively improved after many trains. We also examined the importance of the place control in the evaluation function by the comparison the optimizing results with and without the control of the place and showed the comparison result.
목차
1. Introduction
2. Evaluation Function
3. The Optimization of the Evaluation Function
3.1 Particle Swarm Optimization
3.2 Optimizing Evaluation Function
4. Experimental Results and Discussion
4.1 Particle Swarm Optimization
4.2 Comparison of the Optimizing Results
5. Conclusions and Future Direction
Acknowledgement
References