Rank-based nondomination set identification with preprocessing

Vikas Palakonda, Rammohan Mallipeddi

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

Abstract

In multi-objective optimization, finding a nondomination set is a computationally expensive process and the complexity grows with the popula-tion size. In this paper, we propose a nondomination set identification method with (a) rank-based preprocessing step where the obvious dominant solutions are eliminated and (b) better order of comparison based on the average rank so that the number of comparisons can be significantly reduced. In preprocessing, the maximum rank information of the solutions that are best in each individual objectives is used. In addition, during nondomination set identification process to check if a solution is nondominant the solution is compared only with solutions that are better in terms of average rank. The experiment results demonstrate the effectiveness of the proposed method in identifying the non-dominant set in less number of comparisons.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages150-157
Number of pages8
DOIs
StatePublished - 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9713 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Keywords

  • Multi-objective evolutionary algorithms
  • Nondomination set identification
  • Preprocessing step
  • Rank-based sorting

Fingerprint

Dive into the research topics of 'Rank-based nondomination set identification with preprocessing'. Together they form a unique fingerprint.

Cite this