Hyperbolic Optimizer as a Dynamical System

Simple item page
dc.catalogadoryvc
dc.contributor.authorAlvarado Monardez, Nicolas Felipe Jesús
dc.contributor.authorLobel Díaz, Hans Albert
dc.date.accessioned2025-06-12T16:02:45Z
dc.date.available2025-06-12T16:02:45Z
dc.date.issued2024
dc.description.abstractDuring the last few years, the field of dynamical systems has been developing innovative tools to study the asymptotic behavior of different optimizers in the context of neural networks. In this work, we redefine an extensively studied optimizer, employing classical techniques from hyperbolic geometry. This new definition is linked to a non-linear differential equation as a continuous limit. Additionally, by utilizing Lyapunov stability concepts, we analyze the asymptotic behavior of its critical points.
dc.description.funderMillennium Institute Foundational Research on Data, Chile (IMFD ANID/Millennium Science Initiative Program Code ICN17 002); National Center of Artificial Intelligence, Chile (CENIA FB210017, Basal ANID).
dc.fuente.origenScopus
dc.identifier.issn2640-3498
dc.identifier.scopusidSCOPUS_ID: 85203810414
dc.identifier.urihttps://repositorio.uc.cl/handle/11534/104655
dc.information.autorucEscuela de Ingeniería; Alvarado Monardez, Nicolas Felipe Jesús; S/I; 204444
dc.information.autorucEscuela de Ingeniería; Lobel Díaz, Hans Albert; 0000-0003-3514-9414; 131278
dc.language.isoen
dc.nota.accesocontenido completo
dc.pagina.final1260
dc.pagina.inicio1243
dc.publisherML Research Press
dc.relation.ispartofInternational Conference on Machine Learning : 41 : Vienna, Austria : 2024
dc.revistaProceedings of Machine Learning Research
dc.rightsacceso restringido
dc.subject.ddc620
dc.subject.deweyIngenieríaes_ES
dc.titleHyperbolic Optimizer as a Dynamical System
dc.typecomunicación de congreso
dc.volumen235
sipa.codpersvinculados204444
sipa.codpersvinculados131278
sipa.trazabilidadSCOPUS;2024-09-22
Files
Collections
Artículos de conferencia