摘要
Supportvectormachines(SVMs)haveshownremarkablesuccessinmanyapplications.However,thenon-smoothfeatureofobjectivefunctionisalimitationinpracticalapplicationofSVMs.Toovercomethisdisadvantage,atwicecontinuouslydifferentiablepiecewise-smoothfunctionisconstructedtosmooththeobjectivefunctionofunconstrainedsupportvectormachine(SVM),anditissuesapiecewise-smoothsupportvectormachine(PWESSVM).Comparingtotheothersmoothapproximationfunctions,thesmoothprecisionhasanobviousimprovement.ThetheoreticalanalysisshowsPWESSVMisgloballyconvergent.Numericalresultsandcomparisonsdemonstratetheclassificationperformanceofouralgorithmisbetterthanothercompetitivebaselines.
出版日期
2013年05月15日(中国期刊网平台首次上网日期,不代表论文的发表时间)