Accelerating IEC and EC searches with elite obtained by dimensionality reduction in regression spaces

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

We propose a method for accelerating interactive evolutionary computation (IEC) and evolutionary computation (EC) searches using elite obtained in one-dimensional spaces and use benchmark functions to evaluate the proposed method. The method projects individuals onto n one-dimensional spaces corresponding to each of the n searching parameter axes, approximates each landscape using interpolation or an approximation method, finds the best coordinate from the approximated shape, obtains the elite by combining the best n found coordinates, and uses the elite for the next generation of the IEC or EC. The advantage of this method is that the elite may be easily obtained thanks to their projection onto each one-dimensional space and there is a higher possibility that the elite individual locates near the global optimum. We compare the proposal with methods for obtaining the landscape in the original search space, and show that our proposed method can significantly save computational time. Experimental evaluations of the technique with differential evolution using a simulated IEC user (Gaussian mixture model with different dimensions) and 34 benchmark functions show that the proposed method substantially accelerates IEC and EC searches.

Original languageEnglish
Pages (from-to)27-40
Number of pages14
JournalEvolutionary Intelligence
Volume6
Issue number1
DOIs
Publication statusPublished - Jun 1 2013

Fingerprint

Interactive Evolutionary Computation
Dimensionality Reduction
Evolutionary Computation
Evolutionary algorithms
Regression
Benchmarking
Benchmark
Gaussian Mixture Model
Global Optimum
Differential Evolution
Experimental Evaluation
Approximation Methods
Search Space
Accelerate
Interpolation
Interpolate
Projection
Evaluate

All Science Journal Classification (ASJC) codes

  • Mathematics (miscellaneous)
  • Computer Vision and Pattern Recognition
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Accelerating IEC and EC searches with elite obtained by dimensionality reduction in regression spaces. / Pei, Yan; Takagi, Hideyuki.

In: Evolutionary Intelligence, Vol. 6, No. 1, 01.06.2013, p. 27-40.

Research output: Contribution to journalArticle

@article{c9cfd584b033492d9f3e46ddfaa9025f,
title = "Accelerating IEC and EC searches with elite obtained by dimensionality reduction in regression spaces",
abstract = "We propose a method for accelerating interactive evolutionary computation (IEC) and evolutionary computation (EC) searches using elite obtained in one-dimensional spaces and use benchmark functions to evaluate the proposed method. The method projects individuals onto n one-dimensional spaces corresponding to each of the n searching parameter axes, approximates each landscape using interpolation or an approximation method, finds the best coordinate from the approximated shape, obtains the elite by combining the best n found coordinates, and uses the elite for the next generation of the IEC or EC. The advantage of this method is that the elite may be easily obtained thanks to their projection onto each one-dimensional space and there is a higher possibility that the elite individual locates near the global optimum. We compare the proposal with methods for obtaining the landscape in the original search space, and show that our proposed method can significantly save computational time. Experimental evaluations of the technique with differential evolution using a simulated IEC user (Gaussian mixture model with different dimensions) and 34 benchmark functions show that the proposed method substantially accelerates IEC and EC searches.",
author = "Yan Pei and Hideyuki Takagi",
year = "2013",
month = "6",
day = "1",
doi = "10.1007/s12065-013-0088-9",
language = "English",
volume = "6",
pages = "27--40",
journal = "Evolutionary Intelligence",
issn = "1864-5909",
publisher = "Springer Verlag",
number = "1",

}

TY - JOUR

T1 - Accelerating IEC and EC searches with elite obtained by dimensionality reduction in regression spaces

AU - Pei, Yan

AU - Takagi, Hideyuki

PY - 2013/6/1

Y1 - 2013/6/1

N2 - We propose a method for accelerating interactive evolutionary computation (IEC) and evolutionary computation (EC) searches using elite obtained in one-dimensional spaces and use benchmark functions to evaluate the proposed method. The method projects individuals onto n one-dimensional spaces corresponding to each of the n searching parameter axes, approximates each landscape using interpolation or an approximation method, finds the best coordinate from the approximated shape, obtains the elite by combining the best n found coordinates, and uses the elite for the next generation of the IEC or EC. The advantage of this method is that the elite may be easily obtained thanks to their projection onto each one-dimensional space and there is a higher possibility that the elite individual locates near the global optimum. We compare the proposal with methods for obtaining the landscape in the original search space, and show that our proposed method can significantly save computational time. Experimental evaluations of the technique with differential evolution using a simulated IEC user (Gaussian mixture model with different dimensions) and 34 benchmark functions show that the proposed method substantially accelerates IEC and EC searches.

AB - We propose a method for accelerating interactive evolutionary computation (IEC) and evolutionary computation (EC) searches using elite obtained in one-dimensional spaces and use benchmark functions to evaluate the proposed method. The method projects individuals onto n one-dimensional spaces corresponding to each of the n searching parameter axes, approximates each landscape using interpolation or an approximation method, finds the best coordinate from the approximated shape, obtains the elite by combining the best n found coordinates, and uses the elite for the next generation of the IEC or EC. The advantage of this method is that the elite may be easily obtained thanks to their projection onto each one-dimensional space and there is a higher possibility that the elite individual locates near the global optimum. We compare the proposal with methods for obtaining the landscape in the original search space, and show that our proposed method can significantly save computational time. Experimental evaluations of the technique with differential evolution using a simulated IEC user (Gaussian mixture model with different dimensions) and 34 benchmark functions show that the proposed method substantially accelerates IEC and EC searches.

UR - http://www.scopus.com/inward/record.url?scp=84878574771&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878574771&partnerID=8YFLogxK

U2 - 10.1007/s12065-013-0088-9

DO - 10.1007/s12065-013-0088-9

M3 - Article

AN - SCOPUS:84878574771

VL - 6

SP - 27

EP - 40

JO - Evolutionary Intelligence

JF - Evolutionary Intelligence

SN - 1864-5909

IS - 1

ER -