Show simple item record

Ergodicity of the underdamped mean-field Langevin dynamics

hal.structure.identifierLaboratoire de Mathématiques d'Orsay [LMO]
dc.contributor.authorKazeykina, Anna
hal.structure.identifierCEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
dc.contributor.authorRen, Zhenjie
hal.structure.identifierDepartment of mathematics, Chinese University of Hong Kong
dc.contributor.authorTan, Xiaolu
hal.structure.identifierFakultät für Mathematik [Wien]
dc.contributor.authorYang, Junjian
dc.date.accessioned2020-10-08T12:28:41Z
dc.date.available2020-10-08T12:28:41Z
dc.date.issued2020
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/21090
dc.language.isoenen
dc.subjectmean-field Langevin dynamicsen
dc.subject.ddc004en
dc.titleErgodicity of the underdamped mean-field Langevin dynamicsen
dc.typeDocument de travail / Working paper
dc.description.abstractenWe study the long time behavior of an underdamped mean-field Langevin (MFL) equation , and provide a general convergence as well as an exponential convergence rate result under different conditions. The results on the MFL equation can be applied to study the convergence of the Hamiltonian gradient descent algorithm for the overparametrized optimization. We then provide a numerical example of the algorithm to train a generative adversarial networks (GAN).en
dc.identifier.citationpages29en
dc.relation.ispartofseriestitleCahier de recherche CEREMADEen
dc.identifier.urlsitehttps://hal.archives-ouvertes.fr/hal-02908790en
dc.subject.ddclabelInformatique généraleen
dc.identifier.citationdate2020-07
dc.description.ssrncandidatenonen
dc.description.halcandidatenonen
dc.description.readershiprechercheen
dc.description.audienceInternationalen
dc.date.updated2020-10-08T12:25:06Z
hal.author.functionaut
hal.author.functionaut
hal.author.functionaut
hal.author.functionaut


Files in this item

Thumbnail
Name:
Article_MFL_damping.pdf
Size:
503.0Kb
Format:
PDF
View/Open

This item appears in the following Collection(s)

Show simple item record