Multiplicative updates For Non-Negative Kernel SVM

Simple item page
dc.creatorPotluru, Vamsi K.
dc.creatorPlis, Sergey M.
dc.creatorMorup, Morten
dc.creatorCalhoun, Vince D.
dc.creatorLane, Terran
dc.date2009-02-24
dc.date.accessioned2026-07-07T12:46:18Z
dc.date.available2026-07-07T12:46:18Z
dc.descriptionWe present multiplicative updates for solving hard and soft margin support vector machines (SVM) with non-negative kernels. They follow as a natural extension of the updates for non-negative matrix factorization. No additional param- eter setting, such as choosing learning, rate is required. Ex- periments demonstrate rapid convergence to good classifiers. We analyze the rates of asymptotic convergence of the up- dates and establish tight bounds. We test the performance on several datasets using various non-negative kernels and report equivalent generalization errors to that of a standard SVM.
dc.description4 pages, 1 figure, 1 table
dc.identifierhttps://arxiv.org/abs/0902.4228
dc.identifierhttp://arxiv.org/abs/0902.4228
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/221337
dc.subjectMachine Learning
dc.titleMultiplicative updates For Non-Negative Kernel SVM
dc.typetext

Files

Collections

arXiv