论文信息 - Genetic algorithm with dual species

Genetic algorithm with dual species

In this paper, a new genetic algorithm with two species is proposed. Our dual species genetic algorithm (DSGA) composes of two subpopulation that constitute of same size individuals. The subpopulations have different characteristics, such as crossover probability and mutation operator. In one subpopulation, the parents with higher similarity are cross with higher rate; mutate with general mutation operator. So that, the new algorithm can obtains good exploitation ability. In the other subpopulation, the parents with smaller similarity are cross with higher rate; mutate with big mutation rate. So that, the new algorithm can gets good exploration ability. The performance of our DSGA is compare to that of a single population genetic algorithm (SPGA) and Multi-population genetic algorithm with two populations (2PMGA). The experimental results show that the proposed method can gain higher global convergence rate and higher speed.

Li Ming | Li Junhua | L. Ming | Lin Junhua

[1] Heinrich Braun,et al. On Solving Travelling Salesman Problems by Genetic Algorithms , 1990, PPSN.

[2] K. Dejong,et al. An analysis of the behavior of a class of genetic adaptive systems , 1975 .

[3] Riccardo Poli,et al. Parallel genetic algorithm taxonomy , 1999, 1999 Third International Conference on Knowledge-Based Intelligent Information Engineering Systems. Proceedings (Cat. No.99TH8410).

[4] David E. Goldberg,et al. Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[5] Kenneth Alan De Jong,et al. An analysis of the behavior of a class of genetic adaptive systems. , 1975 .

[6] Yanchun Liang,et al. Solving traveling salesman problems by genetic algorithms , 2003 .

[7] Enrique Alba,et al. Parallelism and evolutionary algorithms , 2002, IEEE Trans. Evol. Comput..

[8] John H. Holland,et al. Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .