非標準支持向量機分類誤差分析.pdf

1、湖北大學(xué)碩士學(xué)位論文非標準支持向量機分類誤差分析姓名：武利平申請學(xué)位級別：碩士專業(yè)：應(yīng)用數(shù)學(xué)指導(dǎo)教師：李落清20070501湖北大學(xué)碩士學(xué)位論文Abstract11地statisticallearningtheoryproposedbyVapniketa1isalleffectivelearningtheoryr11忙supportvectormachines(SVMslisthesuccessfulrealizationofthist

2、heory11“trainingdatesinthelearningtheoryaleindependentandidenticallydistributeddraweraccordingtoaunknownprobabilitydistribution，thentOthelossfunctionanderthisdistributiontaRetheexpectedvalue11lisistheideaofthestandardsup

3、portvectormachinesManyrealworldsituations，howeverarenonstandardinthatmodelisnotvalidforthosesituationsItisessentialtOpromotethelearningalgorithmtOthenonstandardsituationsInthisarticle，Westudytheconvergenceofthelearningal

4、gorithmbaseontheclassificationprobleminthenonstandardsituationsSpecially’weanalyzetheuniformconvergenceandtherelativeuniformconvergenceofthenonstandardsupportvectormachinesbyII他alRSoftheTtkhonovregularizationalgofithmand

5、obtainthegeneralizaOonboundsfocusonbinaryclassificationproblemsInthefirstpart，webriefthebasictheoryofnonstandardsupportveotormachines，and0uninethebasisofSVMInthesecondpan’WebriefthebasisofthesupportVcctormachines11“third

6、partistheimportanceofthisarticleInthispartWestuoytheconvergenceoftheregularizationalgorithmbaseonthetheclassificationproblemAtfirstWestudytheerroranalysisofthesvm1normsoftmarginclassifierandobtaintheboundofitsgeneralizat

7、ioni口rorInthesecond，wefurtherstudyonitsrelativeuniformconvergence，andobtainsatisfiedresultThen，inpartfourextensiontOthenonstandardmultieategorysituationsFinally，inpartfive，WCanalyzethemainresultsgotinpreviouspartsandgive