Show simple item record

dc.contributor.authorMoraglio, A
dc.contributor.authorMcDermott, J
dc.date.accessioned2025-03-12T16:11:00Z
dc.date.issued2025
dc.date.updated2025-03-12T15:41:28Z
dc.description.abstractOn the 30th anniversary of Particle Swarm Optimization (PSO), we present a novel integration of Geometric PSO (GPSO) into the Program Trace Optimization (PTO) framework. GPSO extends PSO to diverse representations in a principled manner, while PTO provides automatic representation design for arbitrary problem structures. By specializing GPSO to PTO’s trace representation, we achieve a universal PSO variant applicable out-of-the-box to any representation and problem. We detail the theoretical foundations of this integration and evaluate our approach on diverse optimization problems across multiple representations. Our work demonstrates the power of combining geometric generalizations with automatic representation design, here specifically obtaining a truly general form of PSO that applies to any representation while retaining its original essence.en_GB
dc.identifier.citationApplications of Evolutionary Computation - 28th European Conference, EvoApplications 2025, held as Part of EvoStar 2025, 23-25 April 2025, Trieste, Italy. Awaiting full citation and DOIen_GB
dc.identifier.urihttp://hdl.handle.net/10871/140606
dc.language.isoenen_GB
dc.publisherSpringer Natureen_GB
dc.relation.urlhttps://link.springer.com/book/9783031900648
dc.rights.embargoreasonUnder temporary indefinite embargo pending publication by Springer Nature. No embargo required on publication en_GB
dc.rights© 2025 The author(s). For the purpose of open access, the author has applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising from this submission.en_GB
dc.titleGeometric particle swarm optimization in Program Trace Optimizationen_GB
dc.typeConference paperen_GB
dc.date.available2025-03-12T16:11:00Z
dc.contributor.editorHart, E
dc.identifier.issn0302-9743
exeter.locationTrieste
dc.descriptionThis is the author accepted manuscript.en_GB
dc.identifier.eissn1611-3349
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/en_GB
dcterms.dateAccepted2025-01-10
dcterms.dateSubmitted2024-11-15
rioxxterms.versionAMen_GB
rioxxterms.licenseref.startdate2025-01-10
rioxxterms.typeConference Paper/Proceeding/Abstracten_GB
refterms.dateFCD2025-03-12T15:41:30Z
refterms.versionFCDAM
refterms.panelBen_GB
pubs.name-of-conferenceEvoApplications
exeter.rights-retention-statementNo


Files in this item

This item appears in the following Collection(s)

Show simple item record

© 2025 The author(s). For the purpose of open access, the author has applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising from this submission.
Except where otherwise noted, this item's licence is described as © 2025 The author(s). For the purpose of open access, the author has applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising from this submission.