Abstract
Performance of a genetic algorithm for function optimization, often appeared in real-world applications, depends on its crossover operator strongly. Existing crossover operators are designed for intensive search in certain promising regions. This paper, first, discusses where the promising search regions are on the basis of some assumptions about the fitness landscapes of objective functions and those about a state of a population, and this discussion reveals that existing crossover operators intensively search some of the promising regions but not all of them. Then, this paper designs a new crossover operator for searching all of the promising regions. For utilizing the advantageous features of this crossover operator, a new selection model considering characteristic preservation is also introduced. Several experiments have shown the proposed method has worked effectively on various test functions.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Ono, I., Yamamura, M., Kobayashi, S.: A genetic algorithm with characteristic preservation for function optimization. In: Proc. of the IIZUKA’96, pp. 511–514 (1996)
Ono, I.: Genetic algorithms for optimization taking account of characteristics preservation (in Japanese). PhD thesis, Tokyo Institute of Technology (1997)
Ono, I., Kobayashi, S.: A real-coded genetic algorithm for function optimization using unimodal normal distribution crossover. In: Proc. of the 7th International Conference on Genetic Algorithms, pp. 246–253 (1997)
Eshelman, L.J., Schaffer, J.D.: Real-Coded Genetic Algorithms and Interval-Schemata. In: Foundations of Genetic Algorithms, vol. 2, pp. 187–202. Morgan Kaufman, Washington (1993)
Tsutsui, S., Yamamura, M., Higuchi, T.: Multi-parent recombination with simplex crossover in real coded genetic algorithms. In: Proc. of the Genetic and Evolutionary Computation Conference 1999, pp. 657–664 (1999)
Kita, H., Yamamura, M.: A functional specialization hypothesis for designing genetic algorithms. In: Proc. of 1999, IEEE International Conference on Systems, Man and Cybernetics, pp. 579–584. IEEE Computer Society Press, Los Alamitos (1999)
Kita, H., Ono, I., Kobayashi, S.: Multi parental extension of the unimodal normal distribution crossover for real-coded genetic algorithms. In: Proc. of the 1999 Congress on Evolutionary Computation, pp. 1581–1587 (1999)
Deb, K., Anand, A., Joshi, D.: A computationally efficient evolutionary algorithm for real-parameter optimization. Evolutionary Computation 10(4), 371–395 (2002)
Satoh, H., Yamamura, M., Kobayashi, S.: Minimal generation gap model for GAs considering both exploration and exploitation. In: Proc. of the IIZUKA’96, pp. 494–497 (1996)
Ikeda, K., Kobayashi, S.: UV phenomenon on genetic algorithms, and UV structure hypothesis (in Japanese). Transactions of the Japanese Society for Artificial Intelligence 17(3), 239–246 (2002)
Ikeda, K., Kobayashi, S.: GA based on the UV-structure hypothesis and its application to JSP. In: Proc. of the Sixth International Conference on Parallel Problem Solving from Nature, pp. 273–282 (2000)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer Berlin Heidelberg
About this paper
Cite this paper
Someya, H. (2007). Promising Search Regions of Crossover Operators for Function Optimization. In: Okuno, H.G., Ali, M. (eds) New Trends in Applied Artificial Intelligence. IEA/AIE 2007. Lecture Notes in Computer Science(), vol 4570. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-73325-6_43
Download citation
DOI: https://doi.org/10.1007/978-3-540-73325-6_43
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-73322-5
Online ISBN: 978-3-540-73325-6
eBook Packages: Computer ScienceComputer Science (R0)