Please use this identifier to cite or link to this item: https://accedacris.ulpgc.es/jspui/handle/10553/144429
DC FieldValueLanguage
dc.contributor.authorTrynda, Jan-
dc.contributor.authorMaczuga, Paweł-
dc.contributor.authorGarcía-Castillo, Luis Emilio-
dc.contributor.authorSchaefer, Robert-
dc.contributor.authorWoźniak, Maciej-
dc.contributor.authorOliver-Serra, Albert-
dc.date.accessioned2025-08-04T18:27:32Z-
dc.date.available2025-08-04T18:27:32Z-
dc.date.issued2025-
dc.identifier.issn1877-7503-
dc.identifier.otherScopus-
dc.identifier.otherWoS-
dc.identifier.urihttps://accedacris.ulpgc.es/handle/10553/144429-
dc.description.abstractDespite their flexibility and success in solving partial differential equations, Physics-Informed Neural Networks (PINNs) often suffer from convergence issues, even failing to converge, particularly in problems with steep gradients or localized features. Several remedies have been suggested to solve this problem, but one of the most promising is the dynamical adaptation of the collocation points. This paper explores a novel adaptive sampling method, of a stochastic nature, based on the Adaptive Mesh Refinement used in the Finite Element Method. The error estimates in our refinement algorithm are based on the value of the residual loss function. We tested our method against a variety of 1D and 2D benchmark problems that exhibit steep gradients near certain boundaries, with promising results.-
dc.languageeng-
dc.relation.ispartofJournal Of Computational Science-
dc.sourceJournal of Computational Science [ISSN 1877-7503], v. 91(102684), (Octubre 2025)-
dc.subjectInvestigación-
dc.subject.otherAdvection-Dominated Problems-
dc.subject.otherAlmost-Singular Poisson Problems-
dc.subject.otherPhysics-Informed Neural Networks-
dc.subject.otherResidual-Based Adaptive Sampling-
dc.titleAn h-adaptive collocation method for Physics-Informed Neural Networks-
dc.typeinfo:eu-repo/semantics/Article-
dc.typeArticle-
dc.identifier.doi10.1016/j.jocs.2025.102684-
dc.identifier.scopus105011699224-
dc.identifier.isi001543496300001-
dc.contributor.orcidNO DATA-
dc.contributor.orcidNO DATA-
dc.contributor.orcidNO DATA-
dc.contributor.orcidNO DATA-
dc.contributor.orcidNO DATA-
dc.contributor.orcid0000-0002-5576-5671-
dc.contributor.authorscopusid58419051000-
dc.contributor.authorscopusid57806951800-
dc.contributor.authorscopusid59228155200-
dc.contributor.authorscopusid35563049800-
dc.contributor.authorscopusid57193304764-
dc.contributor.authorscopusid26031779000-
dc.identifier.eissn1877-7511-
dc.identifier.issue102684-
dc.relation.volume91-
dc.investigacionCiencias-
dc.type2Artículo-
dc.contributor.daisngidNo ID-
dc.contributor.daisngidNo ID-
dc.contributor.daisngidNo ID-
dc.contributor.daisngidNo ID-
dc.contributor.daisngidNo ID-
dc.contributor.daisngidNo ID-
dc.description.numberofpages14-
dc.utils.revision-
dc.contributor.wosstandardWOS:Trynda, J-
dc.contributor.wosstandardWOS:Maczuga, P-
dc.contributor.wosstandardWOS:Oliver-Serra, A-
dc.contributor.wosstandardWOS:Garcia-Castillo, LE-
dc.contributor.wosstandardWOS:Schaefer, R-
dc.contributor.wosstandardWOS:Wozniak, M-
dc.date.coverdateOctubre 2025-
dc.identifier.ulpgc-
dc.contributor.buulpgcBU-INF-
dc.description.sjr0,284
dc.description.jcr1,1
dc.description.sjrqQ3
dc.description.jcrqQ3
dc.description.scieSCIE
dc.description.miaricds11,0
item.fulltextCon texto completo-
item.grantfulltextopen-
crisitem.author.deptGIR SIANI: Modelización y Simulación Computacional-
crisitem.author.deptIU Sistemas Inteligentes y Aplicaciones Numéricas-
crisitem.author.deptDepartamento de Matemáticas-
crisitem.author.orcid0000-0002-3783-8670-
crisitem.author.parentorgIU Sistemas Inteligentes y Aplicaciones Numéricas-
crisitem.author.fullNameOliver Serra, Albert-
Appears in Collections:Artículos
Adobe PDF (4,21 MB)
Show simple item record

Google ScholarTM

Check

Altmetric


Share



Export metadata



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