Publication detail

Distribution Endpoint Estimation Assessment for the Use in Metaheuristic Optimization Procedure

HOLEŠOVSKÝ, J.

Original Title

Distribution Endpoint Estimation Assessment for the Use in Metaheuristic Optimization Procedure

English Title

Distribution Endpoint Estimation Assessment for the Use in Metaheuristic Optimization Procedure

Type

journal article in Scopus

Language

en

Original Abstract

Metaheuristic algorithms are often applied to numerous optimization problems, involving large-scale and mixed-integer instances, specifically. In this contribution we discuss some refinements from the extreme value theory to the lately proposed modification of partition-based random search. The partition-based approach performs iterative random sampling at given feasible subspaces in order to exclude the less favourable regions. The quality of particular regions is evaluated according to the promising index of a region. From statistical perspective, determining the promising index is equivalent to the endpoint estimation of a probability distribution induced by the objective function at the sampling subspace. In the following paper, we give a short review of the recent endpoint estimators derived on the basis of extreme value theory, and compare them by simulations. We discuss also the difficulties in their application and suitability of the estimators for various optimization instances.

English abstract

Metaheuristic algorithms are often applied to numerous optimization problems, involving large-scale and mixed-integer instances, specifically. In this contribution we discuss some refinements from the extreme value theory to the lately proposed modification of partition-based random search. The partition-based approach performs iterative random sampling at given feasible subspaces in order to exclude the less favourable regions. The quality of particular regions is evaluated according to the promising index of a region. From statistical perspective, determining the promising index is equivalent to the endpoint estimation of a probability distribution induced by the objective function at the sampling subspace. In the following paper, we give a short review of the recent endpoint estimators derived on the basis of extreme value theory, and compare them by simulations. We discuss also the difficulties in their application and suitability of the estimators for various optimization instances.

Keywords

metaheuristic optimization; endpoind estimation; extreme value; random search; bootstrap; order statistics.

Released

26.06.2018

Publisher

Brno University of Technology, Faculty of Mechanical Engineering, Institute of Automation and Computer Science

Location

Brno, Czech Republic

ISBN

1803-3814

Periodical

Mendel Journal series

Year of study

24

Number

1

State

CZ

Pages from

93

Pages to

100

Pages count

8

URL

Documents

BibTex


@article{BUT148580,
  author="Jan {Holešovský}",
  title="Distribution Endpoint Estimation Assessment for the Use in Metaheuristic Optimization Procedure",
  annote="Metaheuristic algorithms are often applied to numerous optimization problems, involving large-scale and mixed-integer instances, specifically. In this contribution we discuss some refinements from the extreme value theory to the lately proposed modification of partition-based random search. The partition-based approach performs iterative random sampling at given feasible subspaces in order to exclude the less favourable regions. The quality of particular regions is evaluated according to the promising index of a region. From statistical perspective, determining the promising index is equivalent to the endpoint estimation of a probability distribution induced by the objective function at the sampling subspace. In the following paper, we give a short review of the recent endpoint estimators derived on the basis of extreme value theory, and compare them by simulations. We discuss also the difficulties in their application and suitability of the estimators for various optimization instances.",
  address="Brno University of Technology, Faculty of Mechanical Engineering, Institute of Automation and Computer Science",
  chapter="148580",
  doi="10.13164/mendel.2018.1.093",
  howpublished="print",
  institution="Brno University of Technology, Faculty of Mechanical Engineering, Institute of Automation and Computer Science",
  number="1",
  volume="24",
  year="2018",
  month="june",
  pages="93--100",
  publisher="Brno University of Technology, Faculty of Mechanical Engineering, Institute of Automation and Computer Science",
  type="journal article in Scopus"
}