Two-waykernelMatrixpuncturing:towardsresource-efficientPCAandspectralclusteringRomainCouillet12FlorentChatelain1NicolasLeBihan1Abstractbeprocessedinreal-timebeforebeingdiscardedaltogether.Inparalle...
PrivateAlternatingLeastSquares:PracticalPrivateMatrixCompletionwithTighterRatesSteveChien1PrateekJain1WalidKrichene1SteffenRendle1ShuangSong1AbhradeepThakurta1LiZhang1Abstractknownuser-itemratingsm...
MatrixCompletionwithModel-freeWeightingJiayiWang1RaymondK.W.Wong1XiaojunMao2KwunChuenGaryChan3Abstractguptaetal.,2021)andquantumstatetomography(Wang,2013;Caietal.,2016).Matrixcompletionhasbeenpop-I...
MatrixSketchingforSecureCollaborativeMachineLearningMengjiaoZhang1ShusenWang1Abstractclients,eachclientusesabatchoflocaldatatocomputeastochasticgradient,andtheserveraggregatesthestochas-Collaborati...
LearningNoiseTransitionMatrixfromOnlyNoisyLabelsviaTotalVariationRegularizationYivanZhang12GangNiu2MasashiSugiyama21AbstractBackground.Earlystudiesonlearningfromnoisylabelscanbetracedbacktotherando...
Group-SparseMatrixFactorizationforTransferLearningofWordEmbeddingsKanXu1XuanyiZhao1HamsaBastani1OsbertBastani1Abstractthisstrategyhasbeenusedintransferringimageclassifi-cationmodels(Estevaetal.,201...
FasterKernelMatrixAlgebraviaDensityEstimationArtursBackurs1PiotrIndyk2CameronMusco3TalWagner4Abstractetal.(2008)foranoverview.Westudyfastalgorithmsforcomputingfunda-Kernelmethodstypicallyoperateusi...
Discrete-ValuedLatentPreferenceMatrixEstimationwithGraphSideInformationChanghunJo1KangwookLee2Abstracthavestartedtouseasocialgraphtoresolvethecoldstartproblem.Forinstance,Jamali&Ester(2010)providea...
CatastrophicFisherExplosion:EarlyPhaseFisherMatrixImpactsGeneralizationStanisławJastrze˛bski12DevanshArpit3OliverÅstrand2GiancarloKerg4HuanWang3CaimingXiong3RichardSocher3KyunghyunCho25Krzysztof...
AFrameworkforPrivateMatrixAnalysisinSlidingWindowModelJalajUpadhyay1SarvagyaUpadhyay2Abstractanalysis,includingMatrixanalysis,accuratelycanleakpri-vateinformation(Narayanan&Shmatikov,2006).Asare-We...
AScalableSecondOrderMethodforIll-ConditionedMatrixCompletionfromFewSamplesChristianKu¨mmerle1ClaudioMayrinkVerdun2AbstractedgeofΩandPΩ(X0),wherePΩ:Rd1×d2→Rmisthesubsamplingoperatorthatmapsama...
SupervisedQuantileNormalizationforLow-rankMatrixApproximationMarcoCuturi1OlivierTeboul1JonathanNiles-Weed2Jean-PhilippeVert1Abstractmann,2001;Févotte&Idier,2011),binary(Slawskietal.,2013)orevendes...
SchattenNormsinMatrixStreams:HelloSparsity,GoodbyeDimensionVladimirBraverman1RobertKrauthgamer2AdityaKrishnan1RoiSinoff2Abstractnately,inseveralapplicationdomains,inputmatricesareoftenverysparse,me...
ReinforcementLearninginFeatureSpace:MatrixBandit,Kernels,andRegretBoundLinF.Yang1MengdiWang2Abstractplayanactiona∈A,whereSandAarethestateandactionspaces.ThenthesystemtransitionstoanotherstateExplo...
RandomMatrixTheoryProvesthatDeepLearningRepresentationsofGAN-dataBehaveasGaussianMixturesMohamedElAmineSeddik12CosmeLouart13MohamedTamaazousti1RomainCouillet23Abstractontopofitthatareaccurateforthe...
OrdinalNon-negativeMatrixFactorizationforRecommendationOlivierGouvert1ThomasOberlin2CédricFévotte1Abstract2011)whichaimstopredictfutureinteractionsbetweenusersanditemsinordertomakerecommendations...
MedianMatrixCompletion:fromEmbarrassmenttoOptimalityWeidongLiu1XiaojunMao2RaymondK.W.Wong3Abstractflixdatahaslessthan1%ofobservedentriesofaMatrixwitharound5×105rowsand2×104customers.Withtech-Inth...
LearningFactorizedWeightMatrixforJointFilteringXiangyuXu1YongruiMa2WenxiuSun3Abstractetal.,2010;Suetal.,2019),textureremoval(Xuetal.,2011;Lietal.,2019),super-resolution(Xuetal.,2019a),Jointfilterin...
LatentSpaceFactorisationandManipulationviaMatrixSubspaceProjectionXiaoLi12ChenghuaLin2RuizheLi2ChaozhengWang1FrankGuerin3Abstractabsent.Neverthelesswewouldlikeoursystemtosomehowisolatetheexplanator...
GenerativeFlowswithMatrixExponentialChangyiXiao1LigangLiu1Abstractflowsmodels.Therearetwoprinciplesthatshouldbefol-lowed.First,theJacobiandeterminantshouldbecomputedGenerativeflowsmodelsenjoythepro...