Logo do repositório
 
Publicação

An empirical study on the application of KANs for classification

2025-03-03Documento de conferência Acesso aberto
Ver registo simples
dc.contributor.authorCosta, Samuel Sampaio
dc.contributor.authorPato, Matilde
dc.contributor.authorDatia, Nuno
dc.contributor.institutionNOVALincs
dc.date.accessioned2025-05-27T21:35:21Z
dc.date.available2025-05-27T21:35:21Z
dc.date.issued2025-03-03
dc.descriptionPublisher Copyright: © 2024 Copyright held by the owner/author(s).
dc.description.abstractKolmogorov-Arnold Networks (KANs) represent a breakthrough in deep learning, diverging from Multi-Layer Perceptrons (MLPs) by generalizing the Kolmogorov-Arnold representation theorem (KAT) to networks of arbitrary depth and width. This theorem facilitates the decomposition of multivariate functions into constituent one-dimensional elements, with learnable activation functions on weights and the sum operator on nodes. KANs have been shown to exhibit robust performance in function approximation, validated across mathematical, physical, and practical domains such as traffic prediction and medical diagnostics. Our study investigates KANs’ efficacy through comprehensive evaluations on OpenML, Kaggle and UCI datasets, with a focus on enhancing Human Activity Recognition systems. They demonstrate high classification performance compared to conventional machine learning approaches and MLPs. These findings underscore KANs’ potential as scalable, interpretable tools in modern machine learning applications given their favorable neural scaling laws.en
dc.description.versionpublishersversion
dc.description.versionpublished
dc.format.extent7
dc.format.extent361416
dc.identifier.doi10.1145/3704137.3704166
dc.identifier.isbn9798400718014
dc.identifier.otherPURE: 116901682
dc.identifier.otherPURE UUID: 4311e018-0a28-407d-a090-eddf04d6a753
dc.identifier.otherScopus: 105000414854
dc.identifier.urihttp://hdl.handle.net/10362/183535
dc.identifier.urlhttps://www.scopus.com/pages/publications/105000414854
dc.language.isoeng
dc.peerreviewedyes
dc.publisherACM - Association for Computing Machinery
dc.relationinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F04516%2F2020/PT
dc.relationNOVA Laboratory for Computer Science and Informatics
dc.relationLASIGE - Extreme Computing
dc.relationLASIGE - Extreme Computing
dc.subjectClassification
dc.subjectHuman Activity Recognition
dc.subjectKernel function
dc.subjectKolmogorov-Arnold Networks
dc.subjectMultivariate functions
dc.subjectArtificial Intelligence
dc.titleAn empirical study on the application of KANs for classificationen
dc.typeconference object
degois.publication.firstPage308
degois.publication.lastPage314
degois.publication.titleICAAI 2024 - Conference Proceedings of the 2024 8th International Conference on Advances in Artificial Intelligence
degois.publication.title8th International Conference on Advances in Artificial Intelligence, ICAAI 2024
dspace.entity.typePublication
oaire.awardNumberUIDB/04516/2020
oaire.awardNumberUIDB/00408/2020
oaire.awardNumberUIDP/00408/2020
oaire.awardTitleNOVA Laboratory for Computer Science and Informatics
oaire.awardTitleLASIGE - Extreme Computing
oaire.awardTitleLASIGE - Extreme Computing
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F04516%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F00408%2F2020/PT
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDP%2F00408%2F2020/PT
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
oaire.fundingStream6817 - DCRRNI ID
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
project.funder.nameFundação para a Ciência e a Tecnologia
rcaap.rightsopenAccess
relation.isProjectOfPublication6d79e117-1f85-4221-b1a2-1fe8ff2edce6
relation.isProjectOfPublication3eb06293-18be-490b-a68c-2da04a879a11
relation.isProjectOfPublicationd73f71a1-1b91-43fa-a3b8-df544c07753a
relation.isProjectOfPublication.latestForDiscovery3eb06293-18be-490b-a68c-2da04a879a11

Ficheiros

Principais
A mostrar 1 - 1 de 1
Miniatura
Nome:
Costa_S._S._Pato_M._Datia_N._2025_.pdf
Tamanho:
352.95 KB
Formato:
Adobe Portable Document Format
Ver/Abrir

Coleções

Home collection (FCT)