A more efficient selection sche in iSMS-EMOA

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

doi:10.1007/978-3-319-12027-0_30

A more efficient selection sche in iSMS-EMOA

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

Engineering School

Research output: Contribution to journal › Article › peer-review

4 Citations (Scopus)

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.

Original language	English
Pages (from-to)	371-380
Number of pages	10
Journal	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	8864
DOIs	https://doi.org/10.1007/978-3-319-12027-0_30
Publication status	Published - 1 Jan 2014

Keywords

Hypervolume contribution
Multi-objective evolutionary algorithms
Tuning

ASJC Scopus subject areas

Computer Science(all)
Theoretical Computer Science

Access to Document

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

Cite this

@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{\textquoteright}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 = jan,

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",

}

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

AN - SCOPUS:84921348645

SN - 0302-9743

VL - 8864

SP - 371

EP - 380

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

ER -

A more efficient selection sche in iSMS-EMOA

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this