UnsupervisedEmbeddingAdaptationviaEarly-StageFeatureReconstructionforFew-ShotClassifcationDongHoonLee1Sae-YoungChung1Abstractsetting.Theco-existenceoflabeled-andunlabeled-datainthissettingmotivates...
TowardsRigorousInterpretations:aFormalisationofFeatureAttributionDariusAfchar12RomainHennequin1VincentGuigue2Abstractcontrollinganddisentanglingeveryaspectofamodelpre-diction,rangingfromhowthedatai...
TowardUnderstandingtheFeatureLearningProcessofSelf-supervisedContrastiveLearningZixinWen1YuanzhiLi2Abstractcanevenoutperformthoselearnedbysupervisedlearninginseveraldownstreamtasks.Theremakablepote...
TensorProgramsIV:FeatureLearninginInfinite-WidthNeuralNetworksGregYang1EdwardJ.Hu23AbstractFigure1.PCAofWord2VecembeddingsoftopUScitiesandstates,forNTK,width-64,andwidth-∞FeaturelearningnetworksAs...
SparseFeatureSelectionMakesBatchReinforcementLearningMoreSampleEfficientBotaoHao1YaqiDuan2TorLattimore1CsabaSzepesva´ri13MengdiWang21Abstract1.IntroductionThispaperprovidesastatisticalanalysisofhi...
Privacy-PreservingFeatureSelectionwithSecureMultipartyComputationXilingLi1RafaelDowsley2MartineDeCock13Abstractnature.ProtectionofuserdatahasbecomeasignificantconcerninMLmodeldevelopmentanddeployme...
ParameterlessTransductiveFeatureRe-representationforFew-ShotLearningWentaoCui1YuhongGuo12AbstractFSLsolutionsaregenerallydevelopedintwobranches:meta-learningbasedmethodsandfinetuningbasedmeth-Recen...
NeuralFeatureMatchinginImplicit3DRepresentationsYunluChen1BasuraFernando2HakanBilen3ThomasMensink41EfstratiosGavves1AbstractsourceappearanceFeaturefittingmatchingRecently,neuralimplicitfunctionshav...
NecessaryandSufficientConditionsforCausalFeatureSelectioninTimeSerieswithLatentCommonCausesAtalantiA.Mastakouri1Bernhard,Schölkopf12Dominik,Janzing1Abstractdiscoveryofthecausalgraphfromdataislarge...
MarginalContributionFeatureImportance-anAxiomaticApproachforExplainingDataAmnonCatav1BoyangFu2YazeedZoabi3AhuvaWeiss-Meilik4NoamShomron3JasonErnst256SriramSankararaman257RanGilad-Bachrach8Abstract1...
InterpretingandDisentanglingFeatureComponentsofVariousComplexityfromDNNsJieRen1MingjieLi1ZexuLiu1QuanshiZhang12Abstract?(?)=????+????+?????+??Thispaperaimstodefine,visualize,andana...
FeatureClusteringforSupportIdentificationinExtremeRegionsHamidJalalzai12Re´miLeluc1Abstractseveralrecentstudies:inanomalydetection(Roberts,1999;Cliftonetal.,2011;Goixetal.,2016;Thomasetal.,2017),U...
CountSketches,FeatureHashingandtheMedianofThreeKasperGreenLarsen1RasmusPagh2JakubTeˇtek2Abstractmultitasklearning(Weinbergeretal.,2009)—seesec-tion1.5formoredetails.Inthispaper,werevisittheclassi...
ActiveFeatureAcquisitionwithGenerativeSurrogateModelsYangLi1JunierB.Oliva1Abstractparadigmissuccessfulinamultitudeofdomains,itisincon-gruouswiththeexpectationsofmanyreal-worldintelligentManyreal-wo...
UnravelingMeta-Learning:UnderstandingFeatureRepresentationsforFew-ShotTasksMicahGoldblum1StevenReich1LiamFowl1RenkunNi1ValeriiaCherepanova1TomGoldstein1Abstractdownentirelywhenonlyafewdatasamplesar...
ReinforcementLearninginFeatureSpace:MatrixBandit,Kernels,andRegretBoundLinF.Yang1MengdiWang2Abstractplayanactiona∈A,whereSandAarethestateandactionspaces.ThenthesystemtransitionstoanotherstateExplo...
ProblemswithShapley-value-basedexplanationsasFeatureimportancemeasuresI.ElizabethKumar1SureshVenkatasubramanian1CarlosScheidegger2SorelleA.Friedler3Abstractbythemodel,andthegameisthepredictionofthe...
LearningwithGoodFeatureRepresentationsinBanditsandinRLwithaGenerativeModelTorLattimore1CsabaSzepesva´ri23Gelle´rtWeisz1Abstractforlearninginbandits.TheideasbyDuetal.(2019)suggestthattheanswerisal...
LearningwithFeatureandDistributionEvolvableStreamsZhen-YuZhang1PengZhao1YuanJiang1Zhi-HuaZhou1AbstractFeatureSpaceEvolvingInmanyreal-worldapplications,dataarecol-…………Labellectedintheformofastre...
ImplicitRegularizationofRandomFeatureModelsArthurJacot1BerfinS¸ims¸ek12FrancescoSpadaro1Cle´mentHongler1FranckGabriel1Abstractparametersexhibitsaso-called‘double-descent’curveformanymodelsincl...