Comparative evaluations of evolutionary computation with elite obtained in reduced dimensional spaces

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

8 Citations (Scopus)

Abstract

We propose an elite synthesis optimization strategy for accelerating evolutionary computation (EC) searches using elites obtained from a lower dimensional space. The method projects individuals onto n one-dimensional spaces corresponding to each of the n searching parameter axes, approximates each landscape using Lagrange polynomial interpolation or power function least squares approximation, finds the best coordinate for the approximated shape, obtains the elite by combining the best n found coordinates, and uses the elite for the next generation of the EC. The advantage of this method is that the elite may be easily obtained thanks to their projection onto each one-dimensional space and that there is a higher possibility that the elite will be located near the global optimum. We conduct experimental tests to compare our proposed approaches with previous acceleration approaches using differential evolution and ten benchmark functions. The results demonstrate that the proposed method accelerates EC convergence significantly, especially in early generations.

Original languageEnglish
Title of host publicationProceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011
Pages35-40
Number of pages6
DOIs
Publication statusPublished - Dec 1 2011
Event3rd IEEE International Conference on Intelligent Networking and CollaborativeSystems, INCoS 2011 - Fukuoka, Japan
Duration: Nov 30 2011Dec 2 2011

Publication series

NameProceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011

Other

Other3rd IEEE International Conference on Intelligent Networking and CollaborativeSystems, INCoS 2011
CountryJapan
CityFukuoka
Period11/30/1112/2/11

Fingerprint

Evolutionary algorithms
Least squares approximations
Interpolation
Polynomials

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Networks and Communications

Cite this

Pei, Y., & Takagi, H. (2011). Comparative evaluations of evolutionary computation with elite obtained in reduced dimensional spaces. In Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011 (pp. 35-40). [6132776] (Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011). https://doi.org/10.1109/INCoS.2011.66

Comparative evaluations of evolutionary computation with elite obtained in reduced dimensional spaces. / Pei, Yan; Takagi, Hideyuki.

Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011. 2011. p. 35-40 6132776 (Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011).

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

Pei, Y & Takagi, H 2011, Comparative evaluations of evolutionary computation with elite obtained in reduced dimensional spaces. in Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011., 6132776, Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011, pp. 35-40, 3rd IEEE International Conference on Intelligent Networking and CollaborativeSystems, INCoS 2011, Fukuoka, Japan, 11/30/11. https://doi.org/10.1109/INCoS.2011.66
Pei Y, Takagi H. Comparative evaluations of evolutionary computation with elite obtained in reduced dimensional spaces. In Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011. 2011. p. 35-40. 6132776. (Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011). https://doi.org/10.1109/INCoS.2011.66
Pei, Yan ; Takagi, Hideyuki. / Comparative evaluations of evolutionary computation with elite obtained in reduced dimensional spaces. Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011. 2011. pp. 35-40 (Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011).
@inproceedings{b53eb90cc9104f298e57cccd25aab111,
title = "Comparative evaluations of evolutionary computation with elite obtained in reduced dimensional spaces",
abstract = "We propose an elite synthesis optimization strategy for accelerating evolutionary computation (EC) searches using elites obtained from a lower dimensional space. The method projects individuals onto n one-dimensional spaces corresponding to each of the n searching parameter axes, approximates each landscape using Lagrange polynomial interpolation or power function least squares approximation, finds the best coordinate for the approximated shape, obtains the elite by combining the best n found coordinates, and uses the elite for the next generation of the EC. The advantage of this method is that the elite may be easily obtained thanks to their projection onto each one-dimensional space and that there is a higher possibility that the elite will be located near the global optimum. We conduct experimental tests to compare our proposed approaches with previous acceleration approaches using differential evolution and ten benchmark functions. The results demonstrate that the proposed method accelerates EC convergence significantly, especially in early generations.",
author = "Yan Pei and Hideyuki Takagi",
year = "2011",
month = "12",
day = "1",
doi = "10.1109/INCoS.2011.66",
language = "English",
isbn = "9780769545790",
series = "Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011",
pages = "35--40",
booktitle = "Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011",

}

TY - GEN

T1 - Comparative evaluations of evolutionary computation with elite obtained in reduced dimensional spaces

AU - Pei, Yan

AU - Takagi, Hideyuki

PY - 2011/12/1

Y1 - 2011/12/1

N2 - We propose an elite synthesis optimization strategy for accelerating evolutionary computation (EC) searches using elites obtained from a lower dimensional space. The method projects individuals onto n one-dimensional spaces corresponding to each of the n searching parameter axes, approximates each landscape using Lagrange polynomial interpolation or power function least squares approximation, finds the best coordinate for the approximated shape, obtains the elite by combining the best n found coordinates, and uses the elite for the next generation of the EC. The advantage of this method is that the elite may be easily obtained thanks to their projection onto each one-dimensional space and that there is a higher possibility that the elite will be located near the global optimum. We conduct experimental tests to compare our proposed approaches with previous acceleration approaches using differential evolution and ten benchmark functions. The results demonstrate that the proposed method accelerates EC convergence significantly, especially in early generations.

AB - We propose an elite synthesis optimization strategy for accelerating evolutionary computation (EC) searches using elites obtained from a lower dimensional space. The method projects individuals onto n one-dimensional spaces corresponding to each of the n searching parameter axes, approximates each landscape using Lagrange polynomial interpolation or power function least squares approximation, finds the best coordinate for the approximated shape, obtains the elite by combining the best n found coordinates, and uses the elite for the next generation of the EC. The advantage of this method is that the elite may be easily obtained thanks to their projection onto each one-dimensional space and that there is a higher possibility that the elite will be located near the global optimum. We conduct experimental tests to compare our proposed approaches with previous acceleration approaches using differential evolution and ten benchmark functions. The results demonstrate that the proposed method accelerates EC convergence significantly, especially in early generations.

UR - http://www.scopus.com/inward/record.url?scp=84857171672&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857171672&partnerID=8YFLogxK

U2 - 10.1109/INCoS.2011.66

DO - 10.1109/INCoS.2011.66

M3 - Conference contribution

AN - SCOPUS:84857171672

SN - 9780769545790

T3 - Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011

SP - 35

EP - 40

BT - Proceedings - 3rd IEEE International Conference on Intelligent Networking and Collaborative Systems, INCoS 2011

ER -