Please use this identifier to cite or link to this item: http://hdl.handle.net/1893/36287
Full metadata record
DC FieldValueLanguage
dc.contributor.authorAdair, Jasonen_UK
dc.contributor.authorThomson, Sarah Len_UK
dc.contributor.authorBrownlee, Alexander E Ien_UK
dc.date.accessioned2024-10-08T00:02:58Z-
dc.date.available2024-10-08T00:02:58Z-
dc.date.issued2024-08-01en_UK
dc.identifier.urihttp://hdl.handle.net/1893/36287-
dc.description.abstractWe analyse tness landscapes of evolutionary feature selection to obtain information about feature importance in supervised machine learning. Local optima networks (LONs) are a compact representation of a landscape, and can potentially be adapted for use in explainable artiicial intelligence (XAI). This work examines their applicability for discerning feature importance in supervised machine learning datasets. We visualise aspects of feature selection LONs for a breast cancer prediction dataset as case study, and this process reveals information about the composition of feature sets for the underlying ML models. The estimations of feature importance obtained from LONs are compared with the coeecients extracted from logistic regression models (interpretable AI), and also against feature importances obtained through an established XAI technique: SHAP (explainable AI). We nd that the features present in the LON are not strongly correlated with the model coeecients and SHAP values derived from a model trained prior to feature selection, nor are they strongly correlated within similar groups of local optima after feature selection, calling into question the eeects of constraining the feature space for wrapper-based techniques based on such ranking metrics.en_UK
dc.language.isoenen_UK
dc.publisherACMDLen_UK
dc.relationAdair J, Thomson SL & Brownlee AEI (2024) Explaining evolutionary feature selection via local optima networks. In: GECCO '24 Companion: Genetic and Evolutionary Computation Conference Companion, Melbourne, Australia, 14.07.2024-18.05.2024. ACMDL. https://doi.org/10.1145/3638530.3664183en_UK
dc.rightsThis work is licensed under a Creative Commons Attribution International 4.0 License. GECCO ’24 Companion, July 14–18, 2024, Melbourne, VIC, Australia © 2024 Copyright held by the owner/author(s)en_UK
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/en_UK
dc.subjectCCS CONCEPTSen_UK
dc.subjectMathematics of computingen_UK
dc.subjectGraph algorithmsen_UK
dc.subjectCombina- torial algorithmsen_UK
dc.subjectTheory of computation → Evolutionary algorithms KEYWORDS Fitness Landscapesen_UK
dc.subjectExplainable AIen_UK
dc.subjectLocal Optima Networks (LONs)en_UK
dc.titleExplaining evolutionary feature selection via local optima networksen_UK
dc.typeConference Paperen_UK
dc.identifier.doi10.1145/3638530.3664183en_UK
dc.citation.publicationstatusPublisheden_UK
dc.type.statusVoR - Version of Recorden_UK
dc.author.emailalexander.brownlee@stir.ac.uken_UK
dc.citation.conferencedates2024-07-14 - 2024-05-18en_UK
dc.citation.conferencelocationMelbourne, Australiaen_UK
dc.citation.conferencenameGECCO '24 Companion: Genetic and Evolutionary Computation Conference Companionen_UK
dc.citation.date01/08/2024en_UK
dc.contributor.affiliationComputing Scienceen_UK
dc.contributor.affiliationEdinburgh Napier Universityen_UK
dc.contributor.affiliationComputing Science and Mathematics - Divisionen_UK
dc.identifier.scopusid2-s2.0-85201973695en_UK
dc.identifier.wtid2009526en_UK
dc.contributor.orcid0000-0003-2892-5059en_UK
dc.date.accepted2024-05-03en_UK
dcterms.dateAccepted2024-05-03en_UK
dc.date.filedepositdate2024-09-03en_UK
rioxxterms.apcpaiden_UK
rioxxterms.typeConference Paper/Proceeding/Abstracten_UK
rioxxterms.versionVoRen_UK
local.rioxx.authorAdair, Jason|en_UK
local.rioxx.authorThomson, Sarah L|en_UK
local.rioxx.authorBrownlee, Alexander E I|0000-0003-2892-5059en_UK
local.rioxx.projectInternal Project|University of Stirling|https://isni.org/isni/0000000122484331en_UK
local.rioxx.freetoreaddate2024-10-07en_UK
local.rioxx.licencehttp://creativecommons.org/licenses/by/4.0/|2024-10-07|en_UK
local.rioxx.filenamewksp216s2-file1.pdfen_UK
local.rioxx.filecount1en_UK
Appears in Collections:Computing Science and Mathematics Conference Papers and Proceedings

Files in This Item:
File Description SizeFormat 
wksp216s2-file1.pdfFulltext - Published Version1.91 MBAdobe PDFView/Open


This item is protected by original copyright



A file in this item is licensed under a Creative Commons License Creative Commons

Items in the Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

The metadata of the records in the Repository are available under the CC0 public domain dedication: No Rights Reserved https://creativecommons.org/publicdomain/zero/1.0/

If you believe that any material held in STORRE infringes copyright, please contact library@stir.ac.uk providing details and we will remove the Work from public display in STORRE and investigate your claim.