Deakin University
Browse

File(s) under permanent embargo

An improved scalarization-based dominance evolutionary algorithm for many-objective optimization

Many-objective optimization problems (MaOPs) pose a multitude of challenges for existing multi-objective evolutionary algorithms. One of the key challenges is the poor selection pressure for optimization problems involving a high-dimensional objective space. To overcome this challenge, this paper extends the scalarization-based dominance evolutionary algorithm (SDEA) to improve its convergence rate. Inspired by the neighborhood information sharing scheme between the subproblems in the decomposition-based multi-objective evolutionary algorithm (MOEA/D), a selection mechanism is proposed for enhancing the SDEA in tackling MaOPs. The improved SDEA model is evaluated using different MaOP instances, which include DTLZ and WFG. The results indicate the effectiveness of the enhanced SDEA model in undertaking MaOPs.

History

Event

IEEE Systems Council. Conference (13th : 2019 : Orlando, Fla.)

Series

IEEE Systems Council Conference

Pagination

1 - 5

Publisher

Institute of Electrical and Electronics Engineers

Location

Orlando, Fla.

Place of publication

Piscataway, N.J.

Start date

2019-04-08

End date

2019-04-11

ISBN-13

9781538683965

Language

eng

Publication classification

E1 Full written paper - refereed

Editor/Contributor(s)

[Unknown]

Title of proceedings

SysCon 2019 : Proceedings of the 13th Annual IEEE International Systems Conference

Usage metrics

    Research Publications

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC