A more efficient selection sche in iSMS-EMOA

Adriana Menchaca-Mendez, Elizabeth Montero, María Cristina Riff, Carlos A Coello Coello

Resultado de la investigación: Article

3 Citas (Scopus)

Resumen

In this paper, we study iSMS-EMOA, a recently proposed approach that improves the well-known S metric selection Evolutionary Multi-Objective Algorithm (SMS-EMOA). These two indicator-based multi-objective evolutionary algorithms rely on hypervolume contributions to select individuals. Here, we propose to define a probability of using a randomly selected individual within the iSMS-EMOA’s selection scheme. In order to calibrate the value of such probability, we use the EVOCA tuner. Our preliminary results indicate that we are able to save up to 33% of computations of the contribution to hypervolume with respect to the original iSMS-EMOA, without any significant quality degradation in the solutions obtained. In fact, in some cases, the approach proposed here was even able to improve the quality of the solutions obtained by the original iSMS-EMOA.

Idioma originalEnglish
Páginas (desde-hasta)371-380
Número de páginas10
PublicaciónLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volumen8864
DOI
EstadoPublished - 1 ene 2014

Huella dactilar

Multi-objective Evolutionary Algorithm
Evolutionary algorithms
Degradation
Metric

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Citar esto

@article{556992de6f194e288fb7f84393b6588d,
title = "A more efficient selection sche in iSMS-EMOA",
abstract = "In this paper, we study iSMS-EMOA, a recently proposed approach that improves the well-known S metric selection Evolutionary Multi-Objective Algorithm (SMS-EMOA). These two indicator-based multi-objective evolutionary algorithms rely on hypervolume contributions to select individuals. Here, we propose to define a probability of using a randomly selected individual within the iSMS-EMOA’s selection scheme. In order to calibrate the value of such probability, we use the EVOCA tuner. Our preliminary results indicate that we are able to save up to 33{\%} of computations of the contribution to hypervolume with respect to the original iSMS-EMOA, without any significant quality degradation in the solutions obtained. In fact, in some cases, the approach proposed here was even able to improve the quality of the solutions obtained by the original iSMS-EMOA.",
keywords = "Hypervolume contribution, Multi-objective evolutionary algorithms, Tuning",
author = "Adriana Menchaca-Mendez and Elizabeth Montero and Riff, {Mar{\'i}a Cristina} and Coello, {Carlos A Coello}",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/978-3-319-12027-0_30",
language = "English",
volume = "8864",
pages = "371--380",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",

}

A more efficient selection sche in iSMS-EMOA. / Menchaca-Mendez, Adriana; Montero, Elizabeth; Riff, María Cristina; Coello, Carlos A Coello.

En: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 8864, 01.01.2014, p. 371-380.

Resultado de la investigación: Article

TY - JOUR

T1 - A more efficient selection sche in iSMS-EMOA

AU - Menchaca-Mendez, Adriana

AU - Montero, Elizabeth

AU - Riff, María Cristina

AU - Coello, Carlos A Coello

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In this paper, we study iSMS-EMOA, a recently proposed approach that improves the well-known S metric selection Evolutionary Multi-Objective Algorithm (SMS-EMOA). These two indicator-based multi-objective evolutionary algorithms rely on hypervolume contributions to select individuals. Here, we propose to define a probability of using a randomly selected individual within the iSMS-EMOA’s selection scheme. In order to calibrate the value of such probability, we use the EVOCA tuner. Our preliminary results indicate that we are able to save up to 33% of computations of the contribution to hypervolume with respect to the original iSMS-EMOA, without any significant quality degradation in the solutions obtained. In fact, in some cases, the approach proposed here was even able to improve the quality of the solutions obtained by the original iSMS-EMOA.

AB - In this paper, we study iSMS-EMOA, a recently proposed approach that improves the well-known S metric selection Evolutionary Multi-Objective Algorithm (SMS-EMOA). These two indicator-based multi-objective evolutionary algorithms rely on hypervolume contributions to select individuals. Here, we propose to define a probability of using a randomly selected individual within the iSMS-EMOA’s selection scheme. In order to calibrate the value of such probability, we use the EVOCA tuner. Our preliminary results indicate that we are able to save up to 33% of computations of the contribution to hypervolume with respect to the original iSMS-EMOA, without any significant quality degradation in the solutions obtained. In fact, in some cases, the approach proposed here was even able to improve the quality of the solutions obtained by the original iSMS-EMOA.

KW - Hypervolume contribution

KW - Multi-objective evolutionary algorithms

KW - Tuning

UR - http://www.scopus.com/inward/record.url?scp=84921348645&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-12027-0_30

DO - 10.1007/978-3-319-12027-0_30

M3 - Article

VL - 8864

SP - 371

EP - 380

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -