Complexities of convex combinations and bounding the generalization error in classification

dc.creatorKoltchinskii, Vladimir
dc.creatorPanchenko, Dmitry
dc.date2004-05-18
dc.date2005-08-25
dc.date.accessioned2026-07-07T05:08:23Z
dc.date.available2026-07-07T05:08:23Z
dc.descriptionWe introduce and study several measures of complexity of functions from the convex hull of a given base class. These complexity measures take into account the sparsity of the weights of a convex combination as well as certain clustering properties of the base functions involved in it. We prove new upper confidence bounds on the generalization error of ensemble (voting) classification algorithms that utilize the new complexity measures along with the empirical distributions of classification margins, providing a better explanation of generalization performance of large margin classification methods.
dc.descriptionPublished at http://dx.doi.org/10.1214/009053605000000228 in the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)
dc.identifierhttps://arxiv.org/abs/math/0405356
dc.identifierhttp://arxiv.org/abs/math/0405356
dc.identifierAnnals of Statistics 2005, Vol. 33, No. 4, 1455-1496
dc.identifierdoi:10.1214/009053605000000228
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/71238
dc.subjectProbability
dc.subject62G05 (Primary) 62G20, 60F15 (Secondary)
dc.titleComplexities of convex combinations and bounding the generalization error in classification
dc.typetext

Files

Collections