Improving the performance of the Germinal center artificial immune system using ɛ-dominance: a multi-objective knapsack problem case study

Ayush Joshi*, Christine Zarges, Jonathan Rowe

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

The Germinal center artificial immune system (GC-AIS) is a novel immune algorithm inspired by recent research in immunology, which requires very few parameters to be set by hand. The population of solutions in GC-AIS is dynamic in nature and has no restrictions on its size which can cause problems of population explosion, where the population keeps growing very rapidly, leading to wasteful fitness evaluations. In this paper we try to address this problem in the GC-AIS by incorporating ɛ-dominance, which is a well known mechanism in multiobjective optimization to regulate population size. The improved variant of GC-AIS is compared with a well known multi-objective evolutionary algorithm NSGA-II on the multi-objective knapsack problem. We show that our improved GC-AIS performs better than NSGA-II on the instances of the knapsack problem taken from [23] inheriting the same benefits of having to set fewer parameters manually.

Original languageEnglish
Title of host publicationEvolutionary Computation in Combinatorial Optimization
Subtitle of host publication15th European Conference, EvoCOP 2015, Copenhagen, Denmark, April 8-10, 2015, Proceedings
EditorsGabriela Ochoa, Francisco Chicano
PublisherSpringer
Pages114-125
Number of pages12
Volume9026
ISBN (Electronic)9783319164687
ISBN (Print)9783319164670
DOIs
Publication statusPublished - 2015
Event15th European Conference on Evolutionary Computation in Combinatorial Optimization, EvoCOP 2015 - Copenhagen, Denmark
Duration: 8 Apr 201510 Apr 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9026
ISSN (Print)03029743
ISSN (Electronic)16113349

Conference

Conference15th European Conference on Evolutionary Computation in Combinatorial Optimization, EvoCOP 2015
Country/TerritoryDenmark
CityCopenhagen
Period8/04/1510/04/15

Keywords

  • Artificial immune systems
  • GC-AIS
  • Knapsack problem
  • NSGA-II

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Fingerprint

Dive into the research topics of 'Improving the performance of the Germinal center artificial immune system using ɛ-dominance: a multi-objective knapsack problem case study'. Together they form a unique fingerprint.

Cite this