Abstract

In genetic algorithms(GA), a crossover is performed only at one or two places of a chromosome, and the fixed probabilities of crossover and mutation have been used during the entire generation. A GA with dynamic mutation is known to be superior to GAs with static mutation in performance, but so far no efficient dynamic mutation method has been presented. Accordingly in this paper, a GA is proposed to perform a uniform crossover based on the nucleotide(NU) concept, where DNA and RNA consist of NUs and also a concrete way to vary the probabilities of crossover and mutation dynamically for every generation is proposed. The efficacy of the proposed GA is demonstrated by its application to the unimodal, multimodal and nonlinear control problems, respectively. Simulation results show that in the convergence speed to the optimal value, the proposed GA was superior to existing ones, and the performance of GAs with varying probabilities of the crossover and the mutation improved as compared to GAs with fixed probabilities of the crossover and mutation. And it also shows that the NUs function as the building blocks and so the improvement of the proposed algorithm is supported by the building block hypothesis.

Keywords

• genetic algorithms;
• uniform crossover;
• variable crossover and mutation probabilities;
• building block hypothesis

File

Download PDF

References

1. Adaptation in Natural and Artificial Systems J. H. Holland
2. Genetic Algorithms in Search, Optimization, and Machine Learning D. E. Goldberg
3. IEEE Trans. on Neural Networks v.5 no.1 Convergence analysis of canonical genetic algorithms G. Ruldolph
4. Proc. of 5th Int. Conf. on Machine Learning Experimental results from an evaluation of algorithms that learn to control dynamic systems C. Sammut
5. 제어 · 자동화 · 시스템공학회지 v.1 no.3 인공진화에 의한 학습 및 최적화 장병탁
6. Proc. of 6th Int. Conf. on Genetic Algorithms Adaptive distributed routing using evolutionary fuzzy control B. Carse;T. C. Fogarty;A. Munro
7. IEEE Trans. on Neural Networks v.6 no.2 Combinatorial optimization with use of guided evolutionary simulated annealing P. P.C. Yip;Y. -H. Pao
8. IEEE Trans. on SMC v.23 no.4 A genetics based hybrid scheduler for generating static schedules in flexible manufacturing contexts C. W. Holsapple;V. S. Jacob;R. Pakath;J. S. Zaveri
9. IEEE Trans. on Neural Networks v.5 no.1 Guest editorial evolutionary computation D. B. Fogel;L. J. Fogel
10. GENETIC ALGORITHMS + DATA STRUCTURES = EVOLUTION PROGRAMS Z. Michalewicz
11. A Connectionist Machine for Genetic Hillclimbing D. Ackley
12. Proc. of 3rd Int. Conf. on Genetic Algorithms Uniform crossover in genetic algorithms G. Syswerda
13. Proc. of 4th Int. Conf. on Genetic Algorithms On the virtues of parameterized uniform crossover W. M. Spears;K. A. De Jong
14. Proc. fo 4th Evolutionary Programming Adaptive crossover in evolutionary algorithm W. M. Spears
15. Proc. of 5th Int. Conf. on Genetic Algorithms Optimal mutation rates in genetic search T. Back
16. Proc. of 4th Int. Conf. on Genetic Algorithms An Experimental comparison of binary and floating point representation C. Z. janikow;Z.Michalewitz
17. Proc. of Int. Symposium on Artificial Life and Robots An emergence of fuzzy control rules for mobile robots using DNA coding method T. Yoshikawa;T. Furuhashi;Y. Uchikawa
18. Dynamics of Physical Systems R. H. Cannon Jr.