A Preference-guided Multiobjective Evolutionary Algorithm based on Decomposition

Daniel Edilson de Souza; Fillipe Goulart; Lucas S. Batista; Felipe Campelo

A Preference-guided Multiobjective Evolutionary Algorithm based on Decomposition

Daniel Edilson de Souza, Fillipe Goulart, Lucas S. Batista, Felipe Campelo^*

^*Corresponding author for this work

Research output: Chapter in Book/Published conference output › Conference publication

Abstract

Multiobjective evolutionary algorithms based on decomposition
(MOEA/Ds) represent a class of widely employed problem solvers for multicriteria optimization problems. In this work we investigate the adaptation of
these methods for incorporating preference information prior to the optimization, so that the search process can be biased towards a Pareto-optimal region
that better satisfies the aspirations of a decision-making entity. The incorporation of the Preference-based Adaptive Region-of-interest (PAR) framework into
the MOEA/D requires only the modification of the reference points used within
the scalarization function, which in principle allows a straightforward use in
more sophisticated versions of the base algorithm. Experimental results using
the UF benchmark set suggest gains in diversity within the region of interest,
without significant losses in convergence.

Original language	English
Title of host publication	XIV Encontro Nacional de Inteligencia Artificial e Computacional
Pages	37-48
Publication status	Published - 5 Oct 2017
Event	XIV Encontro Nacional de Inteligência Artificial e Computacional - Uberlândia, Brazil Duration: 2 Oct 2017 → 5 Oct 2017

Conference

Conference	XIV Encontro Nacional de Inteligência Artificial e Computacional
Country/Territory	Brazil
City	Uberlândia
Period	2/10/17 → 5/10/17

Bibliographical note

Access to Document

A Preference-guided Multiobjective Evolutionary Algorithm based on Decomposition
© 2017 The Authors
Accepted author manuscript, 290 KB

Cite this

@inproceedings{11ca332648be45659e7c14e46e13c63e,

title = "A Preference-guided Multiobjective Evolutionary Algorithm based on Decomposition",

abstract = "Multiobjective evolutionary algorithms based on decomposition(MOEA/Ds) represent a class of widely employed problem solvers for multicriteria optimization problems. In this work we investigate the adaptation ofthese methods for incorporating preference information prior to the optimization, so that the search process can be biased towards a Pareto-optimal regionthat better satisfies the aspirations of a decision-making entity. The incorporation of the Preference-based Adaptive Region-of-interest (PAR) framework intothe MOEA/D requires only the modification of the reference points used withinthe scalarization function, which in principle allows a straightforward use inmore sophisticated versions of the base algorithm. Experimental results usingthe UF benchmark set suggest gains in diversity within the region of interest,without significant losses in convergence.",

author = "{de Souza}, {Daniel Edilson} and Fillipe Goulart and Batista, {Lucas S.} and Felipe Campelo",

note = "{\textcopyright} 2017 The Authors; XIV Encontro Nacional de Intelig{\^e}ncia Artificial e Computacional ; Conference date: 02-10-2017 Through 05-10-2017",

year = "2017",

month = oct,

day = "5",

language = "English",

pages = "37--48",

booktitle = "XIV Encontro Nacional de Inteligencia Artificial e Computacional",

}

TY - GEN

T1 - A Preference-guided Multiobjective Evolutionary Algorithm based on Decomposition

AU - de Souza, Daniel Edilson

AU - Goulart, Fillipe

AU - Batista, Lucas S.

AU - Campelo, Felipe

PY - 2017/10/5

Y1 - 2017/10/5

N2 - Multiobjective evolutionary algorithms based on decomposition(MOEA/Ds) represent a class of widely employed problem solvers for multicriteria optimization problems. In this work we investigate the adaptation ofthese methods for incorporating preference information prior to the optimization, so that the search process can be biased towards a Pareto-optimal regionthat better satisfies the aspirations of a decision-making entity. The incorporation of the Preference-based Adaptive Region-of-interest (PAR) framework intothe MOEA/D requires only the modification of the reference points used withinthe scalarization function, which in principle allows a straightforward use inmore sophisticated versions of the base algorithm. Experimental results usingthe UF benchmark set suggest gains in diversity within the region of interest,without significant losses in convergence.

AB - Multiobjective evolutionary algorithms based on decomposition(MOEA/Ds) represent a class of widely employed problem solvers for multicriteria optimization problems. In this work we investigate the adaptation ofthese methods for incorporating preference information prior to the optimization, so that the search process can be biased towards a Pareto-optimal regionthat better satisfies the aspirations of a decision-making entity. The incorporation of the Preference-based Adaptive Region-of-interest (PAR) framework intothe MOEA/D requires only the modification of the reference points used withinthe scalarization function, which in principle allows a straightforward use inmore sophisticated versions of the base algorithm. Experimental results usingthe UF benchmark set suggest gains in diversity within the region of interest,without significant losses in convergence.

UR - http://comissoes.sbc.org.br/ce-ia/pg/historico/base/ENIAC-2017/Anais-ENIAC-2017.pdf

M3 - Conference publication

SP - 37

EP - 48

BT - XIV Encontro Nacional de Inteligencia Artificial e Computacional

T2 - XIV Encontro Nacional de Inteligência Artificial e Computacional

Y2 - 2 October 2017 through 5 October 2017

ER -

A Preference-guided Multiobjective Evolutionary Algorithm based on Decomposition

Abstract

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this