ViLT:Vision-and-LanguageTransformerwithoutConvolutionorRegionSupervisionWonjaeKim1†BokyungSon1IldooKim2AbstractVisualEmbeddingSchemaVision-and-LanguagePre-training(VLP)hasim-RegionFeatureImageCNNR...
UnderstandingSelf-SupervisedLearningDynamicswithoutContrastivePairsYuandongTian1XinleiChen1SuryaGanguli12Abstractquiringexpensivetargetlabels(Devlinetal.,2018).Manystate-of-the-artSSLmethodsincompu...
UCBMomentumQ-learning:CorrectingthebiaswithoutforgettingPierreMénard1OmarDarwicheDomingues2XuedongShang23MichalValko234Abstractbalancetheexplorationoftheenvironmentandexploitationofthecurrentknowl...
ProvablyEnd-to-endLabel-noiseLearningwithoutAnchorPointsXuefengLi12TongliangLiu2BoHan3GangNiu4MasashiSugiyama45Abstract(Arpitetal.,2017;Zhangetal.,2017;Xiaetal.,2021;Wuetal.,2021).Inlabel-noiselear...
PracticalandPrivate(Deep)LearningwithoutSamplingorShufflingPeterKairouz1BrendanMcMahan1ShuangSong1OmThakkar1AbhradeepThakurta1ZhengXu1Abstractinthecontextofdistributedsettingslikefederatedlearn-ing...
ObjectSegmentationwithoutLabelswithLarge-ScaleGenerativeModelsAndreyVoynov1StanislavMorozov1ArtemBabenko1Abstractmodelstoperformlabel-freeobjectsegmentation,wheregroundtruthpixel-levellabelsareexpe...
NeuralArchitectureSearchwithoutTrainingJosephMellor1JackTurner2AmosStorkey2ElliotJ.Crowley3Abstractshiftfromdesigningarchitecturestodesigningalgorithmsthatsearchforcandidatearchitectures(Elskenetal...
MultiplyingMatriceswithoutMultiplyingDavisBlalock12JohnGuttag2AbstractThissettingarisesnaturallyinmachinelearninganddataClassificationminingwhenonehasadatamatrixAwhoserowsaresam-AccuracyMultiplying...
MediatedUncoupledLearning:LearningFunctionswithoutDirectInput-outputCorrespondencesIkkoYamane12JunyaHonda32FlorianYger12MasashiSugiyama24Abstracttionpredictingsentimentofanimage(Mittaletal.,2018).E...
JustTrainTwice:ImprovingGroupRobustnesswithoutTrainingGroupInformationEvanZheranLiu1BehzadHaghgoo1AnnieS.Chen1AditiRaghunathan1PangWeiKoh1ShioriSagawa1PercyLiang1ChelseaFinn1Abstractcanbeespecially...
High-PerformanceLarge-ScaleImageRecognitionwithoutNormalizationAndrewBrock1SohamDe1SamuelL.Smith1KarenSimonyan1AbstractFigure1.ImageNetValidationAccuracyvsTrainingLatency.Allnumbersaresingle-model,...
GlobalRhythmStyleTransferwithoutTextTranscriptionsKaizhiQian12YangZhang12ShiyuChang12JinjunXiong2ChuangGan12DavidCox12MarkHasegawa-Johnson3Abstractwhichthespeakerexpressesthecontent.Therearetwomajo...
FederatedLearningofUserVerificationModelswithoutSharingEmbeddingsHosseinHosseini1HyunsinPark1SungrackYun1ChristosLouizos1JosephSoriaga1MaxWelling1Abstractbeddingiscloseenoughtoareferenceembedding,a...
Distribution-FreeCalibrationGuaranteesforHistogramBinningwithoutSampleSplittingChiragGupta1AadityaRamdas1Abstractthegoalistorecalibrateexisting(‘pre-learnt’)classifiersthatarepowerfuland(statisti...
Weakly-SupervisedDisentanglementwithoutCompromisesFrancescoLocatello12BenPoole3GunnarRätsch1BernhardSchölkopf2OlivierBachem3MichaelTschannen3823>>>>>2Abstract617>>><6661777Intelligentagentsshould...
TwoRoutestoScalableCreditAssignmentwithoutWeightSymmetryDanielKunin1AranNayebi2JavierSagastuy-Brena1SuryaGanguli3JonathanM.Bloom45DanielL.K.Yamins678Abstractforcreditassignmentwithoutweightsymmetry...
TransferLearningwithoutKnowing:ReprogrammingBlack-boxMachineLearningModelswithScarceDataandLimitedResourcesYun-YunTsai1Pin-YuChen2Tsung-YiHo1Abstractatargetdomainbasedontheknowledgetransferredfroma...
ScalableandEfficientComparison-basedSearchwithoutFeaturesDaniyarChumbalov1LucasMaystre2MatthiasGrossglauser1Abstractameaningfulquery.However,thisisoftenanon-trivialtaskforahumanuser.Forexample,most...
NetGANwithoutGAN:FromRandomWalkstoLow-RankApproximationsLucaRendsburg1HolgerHeidrich1UlrikevonLuxburg12Abstractstructure(suchasthecommunitystructure,thecharacteris-ticpathlengths,etc).Beingabletocr...