Un résultat de consistance pour des SVM fonctionnels par interpolation spline

Simple item page
dc.creatorVilla, Nathalie
dc.creatorRossi, Fabrice
dc.date2007-05-02
dc.date.accessioned2026-07-07T07:59:06Z
dc.date.available2026-07-07T07:59:06Z
dc.descriptionThis Note proposes a new methodology for function classification with Support Vector Machine (SVM). Rather than relying on projection on a truncated Hilbert basis as in our previous work, we use an implicit spline interpolation that allows us to compute SVM on the derivatives of the studied functions. To that end, we propose a kernel defined directly on the discretizations of the observed functions. We show that this method is universally consistent.
dc.description6 pages
dc.identifierhttps://arxiv.org/abs/0705.0210
dc.identifierhttp://arxiv.org/abs/0705.0210
dc.identifierComptes Rendus de l Académie des Sciences - Series I - Mathematics 343, 8 (15/10/2006) 555-560
dc.identifierdoi:10.1016/j.crma.2006.09.025
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/128244
dc.subjectStatistics Theory
dc.titleUn résultat de consistance pour des SVM fonctionnels par interpolation spline
dc.typetext

Files

Collections

arXiv