Multi-objective differential evolution with self-navigation

Ke Li, Sam Kwong, Ran Wang, Jingjing Cao, I. Rudas

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Citations (Scopus)

Abstract

Traditional differential evolution (DE) mutation operators explore the search space with no considering the information about the search directions, which results in a purely stochastic behavior. This paper presents a DE variant with self-navigation ability for multi-objective optimization (MODE/SN). It maintains a pool of well designed DE mutation operators with distinct search behaviors and applies them in an adaptive way according to the feedback information from the optimization process. Moreover, we deploy the neural network, which is trained by the extreme learning machine, for mapping an artificially generated solution in the objective space back into the decision space. Empirical results demonstrate that MODE/SN outperforms several state-of-the-art algorithms on a set of benchmark problems with variable linkages.

Original languageEnglish
Title of host publicationConference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
Pages508-513
Number of pages6
DOIs
Publication statusPublished - 2012
Event2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012 - Seoul, Korea, Republic of
Duration: Oct 14 2012Oct 17 2012

Other

Other2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012
CountryKorea, Republic of
CitySeoul
Period10/14/1210/17/12

Fingerprint

Multiobjective optimization
Learning systems
Navigation
Neural networks
Feedback

Keywords

  • Differential evolution
  • multi-objective evolutionary algorithm (MOEA)
  • neural network

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Control and Systems Engineering
  • Human-Computer Interaction

Cite this

Li, K., Kwong, S., Wang, R., Cao, J., & Rudas, I. (2012). Multi-objective differential evolution with self-navigation. In Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics (pp. 508-513). [6377775] https://doi.org/10.1109/ICSMC.2012.6377775

Multi-objective differential evolution with self-navigation. / Li, Ke; Kwong, Sam; Wang, Ran; Cao, Jingjing; Rudas, I.

Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics. 2012. p. 508-513 6377775.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Li, K, Kwong, S, Wang, R, Cao, J & Rudas, I 2012, Multi-objective differential evolution with self-navigation. in Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics., 6377775, pp. 508-513, 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012, Seoul, Korea, Republic of, 10/14/12. https://doi.org/10.1109/ICSMC.2012.6377775
Li K, Kwong S, Wang R, Cao J, Rudas I. Multi-objective differential evolution with self-navigation. In Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics. 2012. p. 508-513. 6377775 https://doi.org/10.1109/ICSMC.2012.6377775
Li, Ke ; Kwong, Sam ; Wang, Ran ; Cao, Jingjing ; Rudas, I. / Multi-objective differential evolution with self-navigation. Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics. 2012. pp. 508-513
@inproceedings{dd589d0fc05a4331bf0ef69765976374,
title = "Multi-objective differential evolution with self-navigation",
abstract = "Traditional differential evolution (DE) mutation operators explore the search space with no considering the information about the search directions, which results in a purely stochastic behavior. This paper presents a DE variant with self-navigation ability for multi-objective optimization (MODE/SN). It maintains a pool of well designed DE mutation operators with distinct search behaviors and applies them in an adaptive way according to the feedback information from the optimization process. Moreover, we deploy the neural network, which is trained by the extreme learning machine, for mapping an artificially generated solution in the objective space back into the decision space. Empirical results demonstrate that MODE/SN outperforms several state-of-the-art algorithms on a set of benchmark problems with variable linkages.",
keywords = "Differential evolution, multi-objective evolutionary algorithm (MOEA), neural network",
author = "Ke Li and Sam Kwong and Ran Wang and Jingjing Cao and I. Rudas",
year = "2012",
doi = "10.1109/ICSMC.2012.6377775",
language = "English",
isbn = "9781467317146",
pages = "508--513",
booktitle = "Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics",

}

TY - GEN

T1 - Multi-objective differential evolution with self-navigation

AU - Li, Ke

AU - Kwong, Sam

AU - Wang, Ran

AU - Cao, Jingjing

AU - Rudas, I.

PY - 2012

Y1 - 2012

N2 - Traditional differential evolution (DE) mutation operators explore the search space with no considering the information about the search directions, which results in a purely stochastic behavior. This paper presents a DE variant with self-navigation ability for multi-objective optimization (MODE/SN). It maintains a pool of well designed DE mutation operators with distinct search behaviors and applies them in an adaptive way according to the feedback information from the optimization process. Moreover, we deploy the neural network, which is trained by the extreme learning machine, for mapping an artificially generated solution in the objective space back into the decision space. Empirical results demonstrate that MODE/SN outperforms several state-of-the-art algorithms on a set of benchmark problems with variable linkages.

AB - Traditional differential evolution (DE) mutation operators explore the search space with no considering the information about the search directions, which results in a purely stochastic behavior. This paper presents a DE variant with self-navigation ability for multi-objective optimization (MODE/SN). It maintains a pool of well designed DE mutation operators with distinct search behaviors and applies them in an adaptive way according to the feedback information from the optimization process. Moreover, we deploy the neural network, which is trained by the extreme learning machine, for mapping an artificially generated solution in the objective space back into the decision space. Empirical results demonstrate that MODE/SN outperforms several state-of-the-art algorithms on a set of benchmark problems with variable linkages.

KW - Differential evolution

KW - multi-objective evolutionary algorithm (MOEA)

KW - neural network

UR - http://www.scopus.com/inward/record.url?scp=84872411104&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872411104&partnerID=8YFLogxK

U2 - 10.1109/ICSMC.2012.6377775

DO - 10.1109/ICSMC.2012.6377775

M3 - Conference contribution

AN - SCOPUS:84872411104

SN - 9781467317146

SP - 508

EP - 513

BT - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics

ER -