Parallel evolutionary algorithms on graphics processing unit

Man Leung WONG, Tien Tsin WONG, Ka Ling FOK

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

45 Citations (Scopus)

Abstract

Evolutionary Algorithms (EAs) are effective and robust methods for solving many practical problems such as feature selection, electrical circuits synthesis, and data mining. However, they may execute for a long time for some difficult problems, because several fitness evaluations must be performed. A promising approach to overcome this limitation is to parallelize these algorithms. In this paper, we propose to implement a parallel EA on consumer-level graphics cards. We perform experiments to compare our parallel EA with an ordinary EA and demonstrate that the former is much more effective than the latter. Since consumer-level graphics cards are available in ubiquitous personal computers and these computers are easy to use and manage, more people will be able to use our parallel algorithm to solve their problems encountered in real-world applications.
Original languageEnglish
Title of host publicationProceedings of the 2005 IEEE Congress on Evolutionary Computation, IEEE CEC 2005
PublisherInstitute of Electrical and Electronics Engineers
Pages2286-2293
Number of pages8
DOIs
Publication statusPublished - 1 Jan 2005

Fingerprint

Parallel algorithms
Evolutionary algorithms
Personal computers
Data mining
Feature extraction
Graphics processing unit
Networks (circuits)
Experiments

Bibliographical note

Paper presented at the 2005 IEEE Congress on Evolutionary Computation (IEEE CEC 2005), 2-5 September 2005, Edinburgh, Scotland.
ISBN of the source publication: 0780393635

Cite this

WONG, M. L., WONG, T. T., & FOK, K. L. (2005). Parallel evolutionary algorithms on graphics processing unit. In Proceedings of the 2005 IEEE Congress on Evolutionary Computation, IEEE CEC 2005 (pp. 2286-2293). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/CEC.2005.1554979
WONG, Man Leung ; WONG, Tien Tsin ; FOK, Ka Ling. / Parallel evolutionary algorithms on graphics processing unit. Proceedings of the 2005 IEEE Congress on Evolutionary Computation, IEEE CEC 2005. Institute of Electrical and Electronics Engineers, 2005. pp. 2286-2293
@inproceedings{7cab84b07532456794fa748602811da3,
title = "Parallel evolutionary algorithms on graphics processing unit",
abstract = "Evolutionary Algorithms (EAs) are effective and robust methods for solving many practical problems such as feature selection, electrical circuits synthesis, and data mining. However, they may execute for a long time for some difficult problems, because several fitness evaluations must be performed. A promising approach to overcome this limitation is to parallelize these algorithms. In this paper, we propose to implement a parallel EA on consumer-level graphics cards. We perform experiments to compare our parallel EA with an ordinary EA and demonstrate that the former is much more effective than the latter. Since consumer-level graphics cards are available in ubiquitous personal computers and these computers are easy to use and manage, more people will be able to use our parallel algorithm to solve their problems encountered in real-world applications.",
author = "WONG, {Man Leung} and WONG, {Tien Tsin} and FOK, {Ka Ling}",
note = "Paper presented at the 2005 IEEE Congress on Evolutionary Computation (IEEE CEC 2005), 2-5 September 2005, Edinburgh, Scotland. ISBN of the source publication: 0780393635",
year = "2005",
month = "1",
day = "1",
doi = "10.1109/CEC.2005.1554979",
language = "English",
pages = "2286--2293",
booktitle = "Proceedings of the 2005 IEEE Congress on Evolutionary Computation, IEEE CEC 2005",
publisher = "Institute of Electrical and Electronics Engineers",

}

WONG, ML, WONG, TT & FOK, KL 2005, Parallel evolutionary algorithms on graphics processing unit. in Proceedings of the 2005 IEEE Congress on Evolutionary Computation, IEEE CEC 2005. Institute of Electrical and Electronics Engineers, pp. 2286-2293. https://doi.org/10.1109/CEC.2005.1554979

Parallel evolutionary algorithms on graphics processing unit. / WONG, Man Leung; WONG, Tien Tsin; FOK, Ka Ling.

Proceedings of the 2005 IEEE Congress on Evolutionary Computation, IEEE CEC 2005. Institute of Electrical and Electronics Engineers, 2005. p. 2286-2293.

Research output: Book Chapters | Papers in Conference ProceedingsConference paper (refereed)Researchpeer-review

TY - GEN

T1 - Parallel evolutionary algorithms on graphics processing unit

AU - WONG, Man Leung

AU - WONG, Tien Tsin

AU - FOK, Ka Ling

N1 - Paper presented at the 2005 IEEE Congress on Evolutionary Computation (IEEE CEC 2005), 2-5 September 2005, Edinburgh, Scotland. ISBN of the source publication: 0780393635

PY - 2005/1/1

Y1 - 2005/1/1

N2 - Evolutionary Algorithms (EAs) are effective and robust methods for solving many practical problems such as feature selection, electrical circuits synthesis, and data mining. However, they may execute for a long time for some difficult problems, because several fitness evaluations must be performed. A promising approach to overcome this limitation is to parallelize these algorithms. In this paper, we propose to implement a parallel EA on consumer-level graphics cards. We perform experiments to compare our parallel EA with an ordinary EA and demonstrate that the former is much more effective than the latter. Since consumer-level graphics cards are available in ubiquitous personal computers and these computers are easy to use and manage, more people will be able to use our parallel algorithm to solve their problems encountered in real-world applications.

AB - Evolutionary Algorithms (EAs) are effective and robust methods for solving many practical problems such as feature selection, electrical circuits synthesis, and data mining. However, they may execute for a long time for some difficult problems, because several fitness evaluations must be performed. A promising approach to overcome this limitation is to parallelize these algorithms. In this paper, we propose to implement a parallel EA on consumer-level graphics cards. We perform experiments to compare our parallel EA with an ordinary EA and demonstrate that the former is much more effective than the latter. Since consumer-level graphics cards are available in ubiquitous personal computers and these computers are easy to use and manage, more people will be able to use our parallel algorithm to solve their problems encountered in real-world applications.

UR - http://commons.ln.edu.hk/sw_master/7109

U2 - 10.1109/CEC.2005.1554979

DO - 10.1109/CEC.2005.1554979

M3 - Conference paper (refereed)

SP - 2286

EP - 2293

BT - Proceedings of the 2005 IEEE Congress on Evolutionary Computation, IEEE CEC 2005

PB - Institute of Electrical and Electronics Engineers

ER -

WONG ML, WONG TT, FOK KL. Parallel evolutionary algorithms on graphics processing unit. In Proceedings of the 2005 IEEE Congress on Evolutionary Computation, IEEE CEC 2005. Institute of Electrical and Electronics Engineers. 2005. p. 2286-2293 https://doi.org/10.1109/CEC.2005.1554979