Publication Details

Elliptical and Archimedean Copulas in Estimation of Distribution Algorithm with Model Migration.

HYRŠ, M.; SCHWARZ, J. Elliptical and Archimedean Copulas in Estimation of Distribution Algorithm with Model Migration. In Proceedings of the 7th International Joint Conference on Computational Intelligence (IJCCI 2015). Lisbon: SciTePress - Science and Technology Publications, 2015. p. 212-219. ISBN: 978-989-758-157-1.
Czech title
Eliptické a Archimedovské kopule v EDA s migrací modelů
Type
conference paper
Language
English
Authors
Hyrš Martin, Ing., Ph.D.
Schwarz Josef, doc. Ing., CSc. (CM-SFE)
Keywords

Estimation of Distribution Algorithms, Copula Theory, Parallel EDA, Island-based Model, Multivariate Copula Sampling, Migration of Probabilistic Models.

Abstract

Estimation of distribution algorithms (EDAs) are stochastic optimization techniques that are based on building and sampling a probability model. Copula theory provides methods that simplify the estimation of a probability model. An island-based version of copula-based EDA with probabilistic model migration (mCEDA) was tested on a set of well-known standard optimization benchmarks in the continuous domain. We investigated two families of copulas - Archimedean and elliptical. Experimental results confirm that this concept of model migration (mCEDA) yields better convergence as compared with the sequential version (sCEDA) and other recently published copula-based EDAs.

Published
2015
Pages
212–219
Proceedings
Proceedings of the 7th International Joint Conference on Computational Intelligence (IJCCI 2015)
ISBN
978-989-758-157-1
Publisher
SciTePress - Science and Technology Publications
Place
Lisbon
EID Scopus
BibTeX
@inproceedings{BUT119927,
  author="Martin {Hyrš} and Josef {Schwarz}",
  title="Elliptical and Archimedean Copulas in Estimation of Distribution Algorithm with Model Migration.",
  booktitle="Proceedings of the 7th International Joint Conference on Computational Intelligence (IJCCI 2015)",
  year="2015",
  pages="212--219",
  publisher="SciTePress - Science and Technology Publications",
  address="Lisbon",
  isbn="978-989-758-157-1",
  url="https://www.fit.vut.cz/research/publication/11013/"
}
Back to top