Faculty of Information Technology, BUT

Publication Details

Estimation of distribution algorithm with copula probabilistic model: A short introduction.

HYRŠ Martin and SCHWARZ Josef. Estimation of distribution algorithm with copula probabilistic model:a short introduction. In: Proceedings of 20th international conference of softcomputing. Brno: Faculty of Mechanical Engineering BUT, 2014, pp. 71-76. ISBN 978-80-214-4984-8.
Czech title
Algoritmus s pravděpodobnostím modelem využívající kopuli:krátký úvod
Type
conference paper
Language
english
Authors
Hyrš Martin, Ing. (DCSY FIT BUT)
Schwarz Josef, doc. Ing., CSc. (DCSY FIT BUT)
Keywords

Estimation of distribution algorithms, Copula Theory, Sklar's theorem, Gaussian copula, optimization problems.

Abstract

A new approach of probabilistic modeling used in Estimation of Distribution Algorithms (EDAs) based on Copula theory is described. By means of copulas it is possible to separate the structure of dependence from the marginal distributions in a joint distribution. Two- dimensional Gaussian copula is depicted in more details including the sampling of the copula based on the conditional probabillity density function. The use of copulas for modeling joint distributions in EDAs is illustrated on several benchmarks.

Published
2014
Pages
71-76
Proceedings
Proceedings of 20th international conference of softcomputing
Conference
20th International Conference on Soft Computing, Brno, CZ
ISBN
978-80-214-4984-8
Publisher
Faculty of Mechanical Engineering BUT
Place
Brno, CZ
BibTeX
@INPROCEEDINGS{FITPUB10678,
   author = "Martin Hyr\v{s} and Josef Schwarz",
   title = "Estimation of distribution algorithm with copula probabilistic model: A short introduction.",
   pages = "71--76",
   booktitle = "Proceedings of 20th international conference of softcomputing",
   year = 2014,
   location = "Brno, CZ",
   publisher = "Faculty of Mechanical Engineering BUT",
   ISBN = "978-80-214-4984-8",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/10678"
}
Files
Back to top