Evolutionary Computing based Analysis of Diversity in Grammatical Evolution

Ayman Youssef, Krishn Kumar Gupt, Muhammad Adil Raja, Aidan Murphy, Conor Ryan

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Diversity is a much sought after aspect of any evolutionary system. More diversity means a cornucopia of diverse behaviors and traits among the individuals of a population. Lack of diversity, on the other hand, leads to a stagnant population whose individuals are more or less similar to each other. Subsequently, they fail to produce a variety of offspring. Grammatical Evolution (GE), being an Evolutionary Algorithm (EA), is also an aspirant of diversity. It allows a GE system to maintain a dynamic population over multiple generations.In this paper, we present our reflections about diversity estimates in a (large) number of experiments. We performed evolutionary experiments to estimate a bunch of well-known benchmark polynomials. We also employed hybrid optimization in our experiments. Our results are insightful. In this paper, we also test the effect of hybrid optimization algorithms integrated with GE on the diversity of the population.

Original languageEnglish
Title of host publicationProceedings - International Conference on Artificial Intelligence and Smart Systems, ICAIS 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1688-1693
Number of pages6
ISBN (Electronic)9781728195377
DOIs
Publication statusPublished - 25 Mar 2021
Event2021 International Conference on Artificial Intelligence and Smart Systems, ICAIS 2021 - Coimbatore, India
Duration: 25 Mar 202127 Mar 2021

Publication series

NameProceedings - International Conference on Artificial Intelligence and Smart Systems, ICAIS 2021

Conference

Conference2021 International Conference on Artificial Intelligence and Smart Systems, ICAIS 2021
Country/TerritoryIndia
CityCoimbatore
Period25/03/2127/03/21

Keywords

  • Diversity
  • Evolutionary computing
  • Grammatical evolution
  • Hybrid optimization
  • Symbolic regression

Fingerprint

Dive into the research topics of 'Evolutionary Computing based Analysis of Diversity in Grammatical Evolution'. Together they form a unique fingerprint.

Cite this