Adaptive, convergent, and diversified archiving strategy for multiobjective evolutionary algorithms

Huidong JIN, Man Leung WONG

Research output: Journal PublicationsJournal Article (refereed)Researchpeer-review

6 Citations (Scopus)

Abstract

It is crucial to obtain automatically and efficiently a well-distributed set of Pareto optimal solutions in multiobjective evolutionary algorithms (MOEAs). Many studies have proposed different evolutionary algorithms that can progress toward the Pareto front with a widely spread distribution of solutions. However, most theoretically, convergent MOEAs necessitate certain prior knowledge about the Pareto front in order to efficiently maintain widespread solutions. In this paper, we propose, based on the new E-dominance concept, an Adaptive Rectangle Archiving (ARA) strategy that overcomes this important problem. The MOEA with this archiving technique provably converges to well-distributed Pareto optimal solutions without prior knowledge about the Pareto front. ARA complements the existing archiving techniques and is useful to both researchers and practitioners.
Original languageEnglish
Pages (from-to)8462-8470
Number of pages9
JournalExpert Systems with Applications
Volume37
Issue number12
DOIs
Publication statusPublished - 1 Dec 2010

Fingerprint

Evolutionary algorithms

Keywords

  • Evolutionary computation
  • convergence
  • diversified archiving
  • multiobjective evolutionary algorithms
  • ϵ-Pareto set

Cite this

@article{a853db7f29c24a9e83edec6baba365a2,
title = "Adaptive, convergent, and diversified archiving strategy for multiobjective evolutionary algorithms",
abstract = "It is crucial to obtain automatically and efficiently a well-distributed set of Pareto optimal solutions in multiobjective evolutionary algorithms (MOEAs). Many studies have proposed different evolutionary algorithms that can progress toward the Pareto front with a widely spread distribution of solutions. However, most theoretically, convergent MOEAs necessitate certain prior knowledge about the Pareto front in order to efficiently maintain widespread solutions. In this paper, we propose, based on the new E-dominance concept, an Adaptive Rectangle Archiving (ARA) strategy that overcomes this important problem. The MOEA with this archiving technique provably converges to well-distributed Pareto optimal solutions without prior knowledge about the Pareto front. ARA complements the existing archiving techniques and is useful to both researchers and practitioners.",
keywords = "Evolutionary computation, convergence, diversified archiving, multiobjective evolutionary algorithms, ϵ-Pareto set",
author = "Huidong JIN and WONG, {Man Leung}",
year = "2010",
month = "12",
day = "1",
doi = "10.1016/j.eswa.2010.05.032",
language = "English",
volume = "37",
pages = "8462--8470",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier Ltd",
number = "12",

}

Adaptive, convergent, and diversified archiving strategy for multiobjective evolutionary algorithms. / JIN, Huidong; WONG, Man Leung.

In: Expert Systems with Applications, Vol. 37, No. 12, 01.12.2010, p. 8462-8470.

Research output: Journal PublicationsJournal Article (refereed)Researchpeer-review

TY - JOUR

T1 - Adaptive, convergent, and diversified archiving strategy for multiobjective evolutionary algorithms

AU - JIN, Huidong

AU - WONG, Man Leung

PY - 2010/12/1

Y1 - 2010/12/1

N2 - It is crucial to obtain automatically and efficiently a well-distributed set of Pareto optimal solutions in multiobjective evolutionary algorithms (MOEAs). Many studies have proposed different evolutionary algorithms that can progress toward the Pareto front with a widely spread distribution of solutions. However, most theoretically, convergent MOEAs necessitate certain prior knowledge about the Pareto front in order to efficiently maintain widespread solutions. In this paper, we propose, based on the new E-dominance concept, an Adaptive Rectangle Archiving (ARA) strategy that overcomes this important problem. The MOEA with this archiving technique provably converges to well-distributed Pareto optimal solutions without prior knowledge about the Pareto front. ARA complements the existing archiving techniques and is useful to both researchers and practitioners.

AB - It is crucial to obtain automatically and efficiently a well-distributed set of Pareto optimal solutions in multiobjective evolutionary algorithms (MOEAs). Many studies have proposed different evolutionary algorithms that can progress toward the Pareto front with a widely spread distribution of solutions. However, most theoretically, convergent MOEAs necessitate certain prior knowledge about the Pareto front in order to efficiently maintain widespread solutions. In this paper, we propose, based on the new E-dominance concept, an Adaptive Rectangle Archiving (ARA) strategy that overcomes this important problem. The MOEA with this archiving technique provably converges to well-distributed Pareto optimal solutions without prior knowledge about the Pareto front. ARA complements the existing archiving techniques and is useful to both researchers and practitioners.

KW - Evolutionary computation

KW - convergence

KW - diversified archiving

KW - multiobjective evolutionary algorithms

KW - ϵ-Pareto set

UR - http://commons.ln.edu.hk/sw_master/173

U2 - 10.1016/j.eswa.2010.05.032

DO - 10.1016/j.eswa.2010.05.032

M3 - Journal Article (refereed)

VL - 37

SP - 8462

EP - 8470

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

IS - 12

ER -