The RBF Hyperparameter in Metamodel-Assisted Evolutionary Search

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

Abstract

RBF metamodels, which are commonly used in expensive optimization problems, rely on a hyperparameter which affects their prediction. The optimal hyperparameter value is typically unknown and hence needs to be estimated by additional procedures. As such this study examines if this overhead is justified from an overall search effectiveness perspective, namely, if changes in the hyperparameter yield significant performance differences. Analysis based on extensive numerical experiments shows that changes are significant in functions with low to moderate multimodality but are less significant in functions with highly multimodality.

Original languageEnglish
Title of host publicationProceedings of the 11th International Conference on Electronics, Communications and Networks, CECNet 2021
EditorsAntonio J. Tallon-Ballesteros
PublisherIOS Press BV
Pages84-89
Number of pages6
ISBN (Electronic)9781643682402
DOIs
StatePublished - 22 Dec 2021
Event11th International Conference on Electronics, Communications and Networks, CECNet 2021 - Virtual, Online, China
Duration: 18 Nov 202121 Nov 2021

Publication series

NameFrontiers in Artificial Intelligence and Applications
Volume345
ISSN (Print)0922-6389
ISSN (Electronic)1879-8314

Conference

Conference11th International Conference on Electronics, Communications and Networks, CECNet 2021
Country/TerritoryChina
CityVirtual, Online
Period18/11/2121/11/21

Keywords

  • Evolutionary algorithms
  • Expensive functions
  • Metamodels
  • RBF

Fingerprint

Dive into the research topics of 'The RBF Hyperparameter in Metamodel-Assisted Evolutionary Search'. Together they form a unique fingerprint.

Cite this