ICSEE 2010, LSMS 2010: Life System Modeling and Intelligent Computing pp 82-88 | Cite as
An Estimation of Distribution Algorithm Based on Clayton Copula and Empirical Margins
Abstract
Estimation of Distribution Algorithms (EDAs) are new evolutionary algorithms which based on the estimation and sampling the distribution model of the selected population in each generation. The way of copula used in EDAs is introduced in this paper. The joint distribution of the selected population is separated into the univariate marginal distribution and a function called copula to represent the dependence structure. And the new individuals are obtained by sampling from copula and then calculating the inverse of the univariate marginal distribution function. The empirical distribution and Clayton copula are used to implement the proposed copula Estimation of Distribution Algorithm (copula EDA). The experimental results show that the proposed algorithm is equivalent to some conventional continuous EDAs in performance.
Keywords
Estimation of distribution algorithms (EDAs) copula theory the joint distribution the marginal distribution Clayton copulaPreview
Unable to display preview. Download preview PDF.
References
- 1.Mühlenbein, H., Paaß, G.: From combination of genes to the estimation of distributions: Binary parameters. In: Ebeling, W., Rechenberg, I., Voigt, H.-M., Schwefel, H.-P. (eds.) PPSN 1996. LNCS, vol. 1141, pp. 178–187. Springer, Heidelberg (1996)CrossRefGoogle Scholar
- 2.Larranaga, P., Lozano, J. (eds.): Estimation of distribution algorithms. A new tool for evolutionary computation. Kluwer Academic Publishers, Dordrecht (2002)Google Scholar
- 3.Pelikan, M.: Hierarchical Bayesian Optimization algorithm: Towards a new generation of evolutionary algorithms. Springer, New York (2005)CrossRefGoogle Scholar
- 4.Shakya, S.: DEUM: A Framework for an Estimation of Distribution Algorithm based on Markov Random Fields. PhD thesis, The Robert Gordon University, Aberdeen, UK (April 2006)Google Scholar
- 5.Duque, T.S., Goldberg, D.E., Sastry, K.: Enhancing the efficiency of the ECGA. In: Rudolph, G., et al. (eds.) PPSN 2008. LNCS, vol. 5199, pp. 165–174. Springer, Heidelberg (2008)CrossRefGoogle Scholar
- 6.Nelsen, R.B. (ed.): An introduction to copulas, 2nd edn. Springer, New York (2006)Google Scholar
- 7.Wang, L.F., Zeng, J.C., Hong, Y.: Estimation of Distribution Algorithm Based on Copula Theory. In: Proceedings of the IEEE Congress on Evolutionary Computation (CEC 2009), Trondheim, Norway, May 18-21, pp. 1057–1063 (2009)Google Scholar
- 8.Salinas-Gutierrez, R., Hernandez-Aguirre, A., Villa-Diharce, E.R.: Using Copulas in Estimation of Distribution Algorithms. In: Hernandez Aguirre, A., et al. (eds.) MICAI 2009. LNCS (LNAI), vol. 5845, pp. 658–668. Springer, Heidelberg (2009)Google Scholar
- 9.Marshall, A.W., Olkin, I.: Families of Multivariate Distributions. Journal of the American statistical association 83, 834–841 (1988)CrossRefGoogle Scholar
- 10.Whelen, N.: Sampling from Archimedean copulas. Quantitative Finance 4(3), 339–352 (2004)CrossRefGoogle Scholar
- 11.Larranaga, P., Etxeberria, R., Lozano, J.A., Pena, J.M.: Optimization in continuous domains by learning and simulation of Gaussian networks. In: Proceedings of the GECCO 2000 Workshop in Optimization by Building and Using Probabilistic Models, pp. 201–204. Morgan Kaufmann, San Francisco (2000)Google Scholar