Modified self-adaptive strategy for controlling parameters in differential evolution

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

Abstract

In this paper, we propose a new technical to modify the self-adaptive Strategy for Controlling Parameters in Differential Evolution algorithm (MSADE). The DE algorithm has been used in many practical cases and has demonstrated good convergence properties. It has only a few control parameters as NP (Number of Particles), F (scaling factor) and CR (crossover), which are kept fixed throughout the entire evolutionary process. However, these control parameters are very sensitive to the setting of the control parameters based on their experiments. The value of control parameters depend on the characteristics of each objective function, so we have to tune their value in each problem that mean it will take too long time to perform. We present a new version of the DE algorithm for obtaining self-adaptive control parameter settings that show good performance on numerical benchmark problems.

Original languageEnglish
Title of host publicationCommunications in Computer and Information Science
Pages370-378
Number of pages9
Volume324 CCIS
EditionPART 2
DOIs
Publication statusPublished - 2012
EventAsia Simulation Conference, AsiaSim 2012 - Shanghai
Duration: 2012 Oct 272012 Oct 30

Publication series

NameCommunications in Computer and Information Science
NumberPART 2
Volume324 CCIS
ISSN (Print)18650929

Other

OtherAsia Simulation Conference, AsiaSim 2012
CityShanghai
Period12/10/2712/10/30

Fingerprint

Experiments

Keywords

  • Differential Evolution (DE)
  • Global search
  • Local search
  • Multi-peak problems

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Tam, B. N., Pham, H., & Hasegawa, H. (2012). Modified self-adaptive strategy for controlling parameters in differential evolution. In Communications in Computer and Information Science (PART 2 ed., Vol. 324 CCIS, pp. 370-378). (Communications in Computer and Information Science; Vol. 324 CCIS, No. PART 2). https://doi.org/10.1007/978-3-642-34390-2_42

Modified self-adaptive strategy for controlling parameters in differential evolution. / Tam, Bui Ngoc; Pham, Hieu; Hasegawa, Hiroshi.

Communications in Computer and Information Science. Vol. 324 CCIS PART 2. ed. 2012. p. 370-378 (Communications in Computer and Information Science; Vol. 324 CCIS, No. PART 2).

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

Tam, BN, Pham, H & Hasegawa, H 2012, Modified self-adaptive strategy for controlling parameters in differential evolution. in Communications in Computer and Information Science. PART 2 edn, vol. 324 CCIS, Communications in Computer and Information Science, no. PART 2, vol. 324 CCIS, pp. 370-378, Asia Simulation Conference, AsiaSim 2012, Shanghai, 12/10/27. https://doi.org/10.1007/978-3-642-34390-2_42
Tam BN, Pham H, Hasegawa H. Modified self-adaptive strategy for controlling parameters in differential evolution. In Communications in Computer and Information Science. PART 2 ed. Vol. 324 CCIS. 2012. p. 370-378. (Communications in Computer and Information Science; PART 2). https://doi.org/10.1007/978-3-642-34390-2_42
Tam, Bui Ngoc ; Pham, Hieu ; Hasegawa, Hiroshi. / Modified self-adaptive strategy for controlling parameters in differential evolution. Communications in Computer and Information Science. Vol. 324 CCIS PART 2. ed. 2012. pp. 370-378 (Communications in Computer and Information Science; PART 2).
@inproceedings{ab7d9451994049ee9bd7509b17db5213,
title = "Modified self-adaptive strategy for controlling parameters in differential evolution",
abstract = "In this paper, we propose a new technical to modify the self-adaptive Strategy for Controlling Parameters in Differential Evolution algorithm (MSADE). The DE algorithm has been used in many practical cases and has demonstrated good convergence properties. It has only a few control parameters as NP (Number of Particles), F (scaling factor) and CR (crossover), which are kept fixed throughout the entire evolutionary process. However, these control parameters are very sensitive to the setting of the control parameters based on their experiments. The value of control parameters depend on the characteristics of each objective function, so we have to tune their value in each problem that mean it will take too long time to perform. We present a new version of the DE algorithm for obtaining self-adaptive control parameter settings that show good performance on numerical benchmark problems.",
keywords = "Differential Evolution (DE), Global search, Local search, Multi-peak problems",
author = "Tam, {Bui Ngoc} and Hieu Pham and Hiroshi Hasegawa",
year = "2012",
doi = "10.1007/978-3-642-34390-2_42",
language = "English",
isbn = "9783642343896",
volume = "324 CCIS",
series = "Communications in Computer and Information Science",
number = "PART 2",
pages = "370--378",
booktitle = "Communications in Computer and Information Science",
edition = "PART 2",

}

TY - GEN

T1 - Modified self-adaptive strategy for controlling parameters in differential evolution

AU - Tam, Bui Ngoc

AU - Pham, Hieu

AU - Hasegawa, Hiroshi

PY - 2012

Y1 - 2012

N2 - In this paper, we propose a new technical to modify the self-adaptive Strategy for Controlling Parameters in Differential Evolution algorithm (MSADE). The DE algorithm has been used in many practical cases and has demonstrated good convergence properties. It has only a few control parameters as NP (Number of Particles), F (scaling factor) and CR (crossover), which are kept fixed throughout the entire evolutionary process. However, these control parameters are very sensitive to the setting of the control parameters based on their experiments. The value of control parameters depend on the characteristics of each objective function, so we have to tune their value in each problem that mean it will take too long time to perform. We present a new version of the DE algorithm for obtaining self-adaptive control parameter settings that show good performance on numerical benchmark problems.

AB - In this paper, we propose a new technical to modify the self-adaptive Strategy for Controlling Parameters in Differential Evolution algorithm (MSADE). The DE algorithm has been used in many practical cases and has demonstrated good convergence properties. It has only a few control parameters as NP (Number of Particles), F (scaling factor) and CR (crossover), which are kept fixed throughout the entire evolutionary process. However, these control parameters are very sensitive to the setting of the control parameters based on their experiments. The value of control parameters depend on the characteristics of each objective function, so we have to tune their value in each problem that mean it will take too long time to perform. We present a new version of the DE algorithm for obtaining self-adaptive control parameter settings that show good performance on numerical benchmark problems.

KW - Differential Evolution (DE)

KW - Global search

KW - Local search

KW - Multi-peak problems

UR - http://www.scopus.com/inward/record.url?scp=84868266772&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84868266772&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-34390-2_42

DO - 10.1007/978-3-642-34390-2_42

M3 - Conference contribution

AN - SCOPUS:84868266772

SN - 9783642343896

VL - 324 CCIS

T3 - Communications in Computer and Information Science

SP - 370

EP - 378

BT - Communications in Computer and Information Science

ER -