简介：一个基于核的判别式分析方法把核称为直接判别式分析被采用，它把直接线性的判别式分析的优点与核诡计的相结合。为了表明它的更好的坚韧性到真实的脸的复杂、非线性的变化，想象，例如照明，面部表情，规模和姿势变化，实验在Olivetti研究实验室，耶鲁和自我造的脸上被执行数据库。Theresults与核主管部件分析和核相对照显示那线性判别式分析，方法能完成更低(7%)用仅仅特征的一个很小的集合的错误率。而且，一个新改正的内核模型被建议改进识别性能。试验性的结果证实它的优势(1%以识别评价)到另外的多项式，核当模特儿。

简介：LetGbeap-seriesgroupandΩbeacompactsubgroupofG.Letλ(x,r)andλ,(x,r)beA-belp-poissontypekernelandproducttypekernelOnΩrespectively.Inthispaperwediscusstheap-proximationpropertiesofsuchkernels,givetheestimate5oftheirmoments,obtainthedirectandin-verseapproximationtheorems.

简介：ＣＯＮＳＩＳＴＥＮＣＹＦＯＲＫＥＲＮＥＬＥＳＴＩＭＡＴＥＯＦＣＯＮＤＩＴＩＯＮＡＬＦＵＮＣＴＩＯＮ￥ＨＵＳｈｕｈｅ（ＤｅｐａｒｔｍｅｎｔＯｆＭａｔｈｅｍｏｔｉｃｓ，ＡｎｈｕｉＵｎｉｖｅｒｓｉｔｙ，Ｈｅｆｅｉ２３００３９，Ｃｈｉｎａ）Ａｂｓｔｒａｃｔ：...

简介：Integralcollisionkerneliselucidatedusingexperimentalresultsfortitania,silicaandaluminananoparticlessynthesizedbyFCVDprocess,andtitaniasubmicronparticlessynthesizedinatubefurnacereactor.Theintegralcollisionkernelwasobtainedfromaparticlenumberbalanceequationbytheintegrationofcollisionratesfromthekinetictheoryofdilutegasesforthefree-moleculeregime,fromtheSmoluchowskitheoryforthecontinuumregime,andbyasemi-empiricalinterpolationforthetransitionregimebetweenthetwolimitingregimes.Comparisonshavebeenmadeonparticlesizeandtheintegralcollisionkernel,showingthatthepredictedintegralcollisionkernelagreedwellwiththeexperimentalresultsinKnudsennumberrangefromabout1.5to20.

简介：Inthispaper,theauthorgivestheweightedweakLipschitzboundednesswithpowerweightforroughmultilinearintegraloperators.Asimplewayisobtainedthatiscloselylinkedwithaclassofroughfractionalintegraloperators.

简介：TheFFDalgorithmisoneofthemostfamousalgorithmsfortheclassicalbinpackingproblem.Inthispaper,someversionsoftheFFDalgorithmareconsideredinseveralbinpackingproblems.Especially,twoofthemappliedtothebinpackingproblemwithkernelitemsareanalyzed.Tightworst-caseperformanceratiosareobtained.

简介：§1.IntroductionandMainResultLet(X,F)beaJBrXR'-valuedvector.AssumethatwhenX=xisgiven,thereexistsaconditionaldensityofYtobedenotedbyf(y[x),whichisaBorel-measurablefunctionof(x,y).Notethatwedonotassumetheexistenceofadensityfunctionof(X,F).Let(X-i,fi),—,(Xn,Fn)bei.i.d.samplesof(X,F).Ourpurposeistoestimatef(y\x)basedonthesesamples.Thisisaninterestingprobleminviewofeitherpuretheoryorpracticalapplications.MotivatedbytheideasuggestedinkernelandNNestimationsinthetheoryofnonparametricregressionanddensityestimates,thefirstauthorproposesthefollowingtwoclassesofestimatorsoff(y\x):

简介：Inthispaper,theauthorsobtainsharpupperandlowerboundsfortheheatkernelassociatedwithJacobitransform,andgetsomeanaloguesofHardy'sTheoremforJacobitransformbyusingthesharpestimateoftheheatkernel.

简介：Amajordifficultyinmultivariablecontroldesignisthecross-couplingbetweeninputsandoutputswhichobscurestheeffectsofaspecificcontrollerontheoverallbehaviorofthesystem.Thispaperconsiderstheapplicationofkernelmethodindecouplingmultivariableoutputfeedbackcontrollers.Simulationresultsarepresentedtoshowthefeasibilityftheproposedtechnique.

简介：

简介：Previously,anovelclassifiercalledKernel-basedNonlinearDiscriminator(KND)wasproposedtodiscriminateapatternclassfromotherclassesbyminimizingmeaneffectofthelatter.Toconsidertheeffectofthetargetclass,thispaperintroducesanobliqueprojectionalgorithmtodeterminethecoefficientsofaKNDsothatitisextendedtoanewversioncalledextendedKND(eKND).IneKNDconstruction,thedesiredoutputvectorofthetargetclassisobliquelyprojectedontotherelevantsubspacealongthesubspacerelatedtootherclasses.Inaddition,asimpletechniqueisproposedtocalculatetheassociatedobliqueprojectionoperator.ExperimentalresultsonhandwrittendigitrecognitionshowthatthealgorithmperformesbetterthanaKNDclassifierandsomeothercommonlyusedclassifiers.

简介：

简介：1.IntroductionTheestimationofpopulationquaillesisofgrestillterestwhenone.isnotpreparedtoassumeaparametricformfortheunderlyingdistribution.Inaddition,quaillesoftenariseasthensturalthingtoestimatewhentheunderlyingdistributionisskewed.LetXIIXZ,’’’)Xubei...

简介：Inthispaper,thenormalapproximationrateandtherandomweightingapproximationrateoferrordistributionofthekernelestimatorofconditionaldensityfunctionf(y!|x)arestudied.Theresultsmaybeusedtoconstructtheconfidenceintervaloff(y|x).

简介：InthispaperwegiveLp-boundednessfortheoperatorTudefinedbywhereP(x,y)isarealnontrivialpolynomialonRn×Rn,Ωishomogeneousofdegreezero,Ω∈Lq(Sn-1),q>1/(1-μ)andb(r)∈BV(R+),TheresultcanberegardedasanimprovementofF.RicciandE.M.Stein’sresultforfractionaloscillatoryintegraloperatorwithsmoothnesskernel.

简介：用本地化的方法，作者与C为一个相对紧缩的领域用Bochner-Martinelli核获得单个积分的排列公式((1))斯坦因上的光滑的边界歧管。作为一个应用程序，作者用Bochner-Martinelli核和可变系数为线性单个不可分的方程讨论规则化问题；用排列公式，单个不可分的方程能被归结为一个fredholm方程。给词调音：斯坦因歧管；有Bochner-Martinelli核的单个积分；排列公式；规则化问题

简介：Inpractice,retrainingatrainedclassifierisnecessarywhennoveldatabecomeavailable.ThispaperadoptsanincrementallearningproceduretoadaptivelytrainaKernel-basedNonlinearRepresentor(KNR),arecentlypresentednonlinearclassifierforoptimalpatternrepresentation,sothatitsgeneralizationabilitymaybeevaluatedintime-variantsituationandasparserrepresentationisobtainedforcomputationallyintensivetasks.Theaddressedtechniquesareappliedtohandwrittendigitclassificationtoillustratethefeasibilityforpatternrecognition.

简介：Basedonlefttruncatedandrightcensoreddependentdata,theestimatorsofhigherderivativesofdensityfunctionandhazardratefunctionaregivenbykernelsmoothingmethod.Whenobserveddataexhibitα-mixingdependence,localpropertiesincludingstrongconsistencyandlawofiteratedlogarithmarepresented.Moreover,whenthemodeestimatorisdefinedastherandomvariablethatmaximizesthekerneldensityestimator,theasymptoticnormalityofthemodeestimatorisestablished.

简介：Theauthorsderivelawsoftheiteratedlogarithmforkernelestimatorofregressionfunctionbasedondirectionaldata.Theresultsaredistributionfreeinthesensethattheyaretrueforalldistributionsofdesignvariable.

简介：Theissueofselectionofbandwidthinkernelsmoothingmethodisconsideredwithinthecontextofpartiallylinearmodels.Inthispaper,westudytheasymptoticbehaviorofthebandwidthchoicebasedongeneralizedcross-validation(GCV)approachandprovethatthisbandwidthchoiceisasymptoticallyoptimal.Numericalsimulationarealsoconductedtoinvestigatetheempiricalperformanceofgeneralizedcross-validation.