An overview of weighted and unconstrained scalarizing functions

Miriam Pescador-Rojas, Raquel Hernández Gómez, Elizabeth Montero, Nicolás Rojas-Morales, María Cristina Riff, Carlos A. Coello Coello

Resultado de la investigación: Conference contribution

10 Citas (Scopus)

Resumen

Scalarizing functions play a crucial role in multi-objective evolutionary algorithms (MOEAs) based on decomposition and the R2 indicator, since they guide the population towards nearly optimal solutions, assigning a fitness value to an individual according to a predefined target direction in objective space. This paper presents a general review of weighted scalarizing functions without constraints, which have been proposed not only within evolutionary multi-objective optimization but also in the mathematical programming literature. We also investigate their scalability up to 10 objectives, using the test problems of Lamé Superspheres on the MOEA/D and MOMBI-II frameworks. For this purpose, the best suited scalarizing functions and their model parameters are determined through the evolutionary calibrator EVOCA. Our experimental results reveal that some of these scalarizing functions are quite robust and suitable for handling many-objective optimization problems.

Idioma originalEnglish
Título de la publicación alojadaEvolutionary Multi-Criterion Optimization - 9th International Conference, EMO 2017, Proceedings
EditoresOliver Schütze, Gunter Rudolph, Kathrin Klamroth, Yaochu Jin, Heike Trautmann, Christian Grimme, Margaret Wiecek
EditorialSpringer Verlag
Páginas499-513
Número de páginas15
ISBN (versión impresa)9783319541563
DOI
EstadoPublished - 1 ene 2017
Evento9th International Conference on Evolutionary Multi-Criterion Optimization, EMO 2017 - Munster, Germany
Duración: 19 mar 201722 mar 2017

Serie de la publicación

NombreLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volumen10173 LNCS
ISSN (versión impresa)0302-9743
ISSN (versión digital)1611-3349

Conference

Conference9th International Conference on Evolutionary Multi-Criterion Optimization, EMO 2017
PaísGermany
CiudadMunster
Período19/03/1722/03/17

Huella dactilar

Multi-objective Evolutionary Algorithm
Evolutionary algorithms
Evolutionary multiobjective Optimization
Mathematical programming
Multiobjective optimization
Mathematical Programming
Fitness
Test Problems
Scalability
Optimal Solution
Optimization Problem
Decomposition
Decompose
Target
Experimental Results
Model
Review
Framework

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Citar esto

Pescador-Rojas, M., Gómez, R. H., Montero, E., Rojas-Morales, N., Riff, M. C., & Coello Coello, C. A. (2017). An overview of weighted and unconstrained scalarizing functions. En O. Schütze, G. Rudolph, K. Klamroth, Y. Jin, H. Trautmann, C. Grimme, & M. Wiecek (Eds.), Evolutionary Multi-Criterion Optimization - 9th International Conference, EMO 2017, Proceedings (pp. 499-513). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10173 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-54157-0_34
Pescador-Rojas, Miriam ; Gómez, Raquel Hernández ; Montero, Elizabeth ; Rojas-Morales, Nicolás ; Riff, María Cristina ; Coello Coello, Carlos A. / An overview of weighted and unconstrained scalarizing functions. Evolutionary Multi-Criterion Optimization - 9th International Conference, EMO 2017, Proceedings. editor / Oliver Schütze ; Gunter Rudolph ; Kathrin Klamroth ; Yaochu Jin ; Heike Trautmann ; Christian Grimme ; Margaret Wiecek. Springer Verlag, 2017. pp. 499-513 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{4fd79e2aace04f6793e047744d7cbf7e,
title = "An overview of weighted and unconstrained scalarizing functions",
abstract = "Scalarizing functions play a crucial role in multi-objective evolutionary algorithms (MOEAs) based on decomposition and the R2 indicator, since they guide the population towards nearly optimal solutions, assigning a fitness value to an individual according to a predefined target direction in objective space. This paper presents a general review of weighted scalarizing functions without constraints, which have been proposed not only within evolutionary multi-objective optimization but also in the mathematical programming literature. We also investigate their scalability up to 10 objectives, using the test problems of Lam{\'e} Superspheres on the MOEA/D and MOMBI-II frameworks. For this purpose, the best suited scalarizing functions and their model parameters are determined through the evolutionary calibrator EVOCA. Our experimental results reveal that some of these scalarizing functions are quite robust and suitable for handling many-objective optimization problems.",
keywords = "Evolutionary algorithms, Many-objective optimization, Scalarizing function, Tuning process",
author = "Miriam Pescador-Rojas and G{\'o}mez, {Raquel Hern{\'a}ndez} and Elizabeth Montero and Nicol{\'a}s Rojas-Morales and Riff, {Mar{\'i}a Cristina} and {Coello Coello}, {Carlos A.}",
year = "2017",
month = "1",
day = "1",
doi = "10.1007/978-3-319-54157-0_34",
language = "English",
isbn = "9783319541563",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "499--513",
editor = "Oliver Sch{\"u}tze and Gunter Rudolph and Kathrin Klamroth and Yaochu Jin and Heike Trautmann and Christian Grimme and Margaret Wiecek",
booktitle = "Evolutionary Multi-Criterion Optimization - 9th International Conference, EMO 2017, Proceedings",
address = "Germany",

}

Pescador-Rojas, M, Gómez, RH, Montero, E, Rojas-Morales, N, Riff, MC & Coello Coello, CA 2017, An overview of weighted and unconstrained scalarizing functions. En O Schütze, G Rudolph, K Klamroth, Y Jin, H Trautmann, C Grimme & M Wiecek (eds.), Evolutionary Multi-Criterion Optimization - 9th International Conference, EMO 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10173 LNCS, Springer Verlag, pp. 499-513, 9th International Conference on Evolutionary Multi-Criterion Optimization, EMO 2017, Munster, Germany, 19/03/17. https://doi.org/10.1007/978-3-319-54157-0_34

An overview of weighted and unconstrained scalarizing functions. / Pescador-Rojas, Miriam; Gómez, Raquel Hernández; Montero, Elizabeth; Rojas-Morales, Nicolás; Riff, María Cristina; Coello Coello, Carlos A.

Evolutionary Multi-Criterion Optimization - 9th International Conference, EMO 2017, Proceedings. ed. / Oliver Schütze; Gunter Rudolph; Kathrin Klamroth; Yaochu Jin; Heike Trautmann; Christian Grimme; Margaret Wiecek. Springer Verlag, 2017. p. 499-513 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10173 LNCS).

Resultado de la investigación: Conference contribution

TY - GEN

T1 - An overview of weighted and unconstrained scalarizing functions

AU - Pescador-Rojas, Miriam

AU - Gómez, Raquel Hernández

AU - Montero, Elizabeth

AU - Rojas-Morales, Nicolás

AU - Riff, María Cristina

AU - Coello Coello, Carlos A.

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Scalarizing functions play a crucial role in multi-objective evolutionary algorithms (MOEAs) based on decomposition and the R2 indicator, since they guide the population towards nearly optimal solutions, assigning a fitness value to an individual according to a predefined target direction in objective space. This paper presents a general review of weighted scalarizing functions without constraints, which have been proposed not only within evolutionary multi-objective optimization but also in the mathematical programming literature. We also investigate their scalability up to 10 objectives, using the test problems of Lamé Superspheres on the MOEA/D and MOMBI-II frameworks. For this purpose, the best suited scalarizing functions and their model parameters are determined through the evolutionary calibrator EVOCA. Our experimental results reveal that some of these scalarizing functions are quite robust and suitable for handling many-objective optimization problems.

AB - Scalarizing functions play a crucial role in multi-objective evolutionary algorithms (MOEAs) based on decomposition and the R2 indicator, since they guide the population towards nearly optimal solutions, assigning a fitness value to an individual according to a predefined target direction in objective space. This paper presents a general review of weighted scalarizing functions without constraints, which have been proposed not only within evolutionary multi-objective optimization but also in the mathematical programming literature. We also investigate their scalability up to 10 objectives, using the test problems of Lamé Superspheres on the MOEA/D and MOMBI-II frameworks. For this purpose, the best suited scalarizing functions and their model parameters are determined through the evolutionary calibrator EVOCA. Our experimental results reveal that some of these scalarizing functions are quite robust and suitable for handling many-objective optimization problems.

KW - Evolutionary algorithms

KW - Many-objective optimization

KW - Scalarizing function

KW - Tuning process

UR - http://www.scopus.com/inward/record.url?scp=85014241346&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-54157-0_34

DO - 10.1007/978-3-319-54157-0_34

M3 - Conference contribution

SN - 9783319541563

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 499

EP - 513

BT - Evolutionary Multi-Criterion Optimization - 9th International Conference, EMO 2017, Proceedings

A2 - Schütze, Oliver

A2 - Rudolph, Gunter

A2 - Klamroth, Kathrin

A2 - Jin, Yaochu

A2 - Trautmann, Heike

A2 - Grimme, Christian

A2 - Wiecek, Margaret

PB - Springer Verlag

ER -

Pescador-Rojas M, Gómez RH, Montero E, Rojas-Morales N, Riff MC, Coello Coello CA. An overview of weighted and unconstrained scalarizing functions. En Schütze O, Rudolph G, Klamroth K, Jin Y, Trautmann H, Grimme C, Wiecek M, editores, Evolutionary Multi-Criterion Optimization - 9th International Conference, EMO 2017, Proceedings. Springer Verlag. 2017. p. 499-513. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-54157-0_34