WILDS:ABenchmarkofin-the-WildDistributionShiftsPangWeiKoh1ShioriSagawa1HenrikMarklund1SangMichaelXie1MarvinZhang2AkshayBalsubramani1WeihuaHu1MichihiroYasunaga1RichardLanasPhillips3IrenaGao1TonyLee1...
OptiDICE:OfflinePolicyOptimizationviaStationaryDistributionCorrectionEstimationJongminLee1WonseokJeon23Byung-JunLee4JoellePineau235Kee-EungKim16Abstractandthentodeploythemodelwithitsparameterfixedw...
Near-OptimalLinearRegressionunderDistributionShiftQiLei1WeiHu1JasonD.Lee1AbstractprovidehelpfulmarginalDistributioninformationaboutthetargetdomain.AlthoughthisapproachisintegraltomanyTransferlearni...
MANDOLINE:ModelEvaluationunderDistributionShiftMayeeChen1KaranGoel1NimitSohoni2FaitPoms1KayvonFatahalian1ChristopherRe´1Abstracttionerstodetermineiftheirmodelswillperformwellwhendeployed.Unfortuna...
LabelDistributionLearningMachineJingWang12XinGeng12AbstractHAP:1.39descriptiondegree0.30SAD:2.030.25AlthoughLabelDistributionLearning(LDL)hasSUR:1.57witnessedextensiveclassificationapplications,itA...
ExaminingandCombatingSpuriousFeaturesunderDistributionShiftChuntingZhou1XuezheMa2PaulMichel1GrahamNeubig1Abstract(a)ERM(b)GroupDRO[clean](average=93.3%robust=83.4%)(average=96.5%robust=95.8%)Acentr...
DiffusionEarthMover’sDistanceandDistributionEmbeddingsAlexanderTong1GuillaumeHuguet23AmineNatik23KincaidMacDonald4ManikKuchroo51RonaldR.Coifman4GuyWolf†23SmitaKrishnaswamy†51Abstract1.Introducti...
ApproximatingaDistributionUsingWeightQueriesNadavBarak1SivanSabato1Abstractinterest.However,inmanycases,obtainingsucharan-domsampleisdifficultorimpossible.Inthiswork,weWeconsideranovelchallenge:app...
AStructuredObservationDistributionforGenerativeBiologicalSequencePredictionandForecastingEliN.Weinstein1DeboraS.Marks23Abstractsequencedata,makingpredictionsaboutunobservedorfu-turesequences,andgen...
Thek-tiedNormalDistribution:ACompactParameterizationofGaussianMeanFieldPosteriorsinBayesianNeuralNetworksJakubSwiatkowski1+KevinRoth2+BastiaanS.Veeling34+LinhTran5+JoshuaV.Dillon4JasperSnoek4Stepha...
TheEffectofNaturalDistributionShiftonQuestionAnsweringModelsJohnMiller1KarlKrauth1BenjaminRecht1LudwigSchmidt1AbstractmerelytoobtainhighscoresontheSQuADleaderboard,butrathertogeneralizetonewexample...
Test-TimeTrainingwithSelf-SupervisionforGeneralizationunderDistributionShiftsYuSun1XiaolongWang12ZhuangLiu1JohnMiller1AlexeiA.Efros1MoritzHardt1AbstractOurproposedtest-timetrainingmethodcreatesasel...
RobustandStableBlackBoxExplanationsHimabinduLakkaraju1NinoArsov2OsbertBastani3Abstractanddecidewhenandhowtoemploythem.However,theproprietarynatureandincreasingcomplexityofmachineAsmachinelearningbl...
NADS:NeuralArchitectureDistributionSearchforUncertaintyAwarenessRandyArdywibowo1ShahinBoluki1XinyuGong2ZhangyangWang2XiaoningQian1Abstractlousdatacancomeinsettingssuchasinautonomousdriv-ing(Kendall...
LearningwithFeatureandDistributionEvolvableStreamsZhen-YuZhang1PengZhao1YuanJiang1Zhi-HuaZhou1AbstractFeatureSpaceEvolvingInmanyreal-worldapplications,dataarecol-…………Labellectedintheformofastre...
ConstructiveUniversalHigh-DimensionalDistributionGenerationthroughDeepReLUNetworksDmytroPerekrestenko1StephanMu¨ller1HelmutBo¨lcskei12Abstractquence.Here,theinputlayersizeisdeterminedbythedimensi...
CharacterizingDistributionEquivalenceandStructureLearningforCyclicandAcyclicDirectedGraphsAmirEmadGhassami1AlanYang1NegarKiyavash2KunZhang3Abstractofthedata(Koller&Friedman,2009).Wheninterpretedcau...
BatchStationaryDistributionEstimationJunfengWen1BoDai2LihongLi2DaleSchuurmans12Abstractunderlyingprocess.Nevertheless,onewouldstillliketoestimatetargetpropertiesofthestationaryDistribution,suchWeco...
Multivariate-InformationAdversarialEnsembleforScalableJointDistributionMatchingZiliangChen1ZhanfuYang2XiaoxiWang1XiaodanLiang1XiaopengYan1GuanbinLi1LiangLin1Abstract(∀j∈[m],j=i)tosatisfytheequati...
Doubly-CompetitiveDistributionEstimationYiHao1AlonOrlitsky1Abstract1.1.DistributionEstimationDistributionestimationisastatistical-learningLet∆kdenotethecollectionofDistributionsoverthedis-cornerst...
