WatermarkingDeepNeuralNetworkswithGreedyResidualsHanwenLiu1ZhenyuWeng1YueshengZhu1Abstractandprivacyissues(Shokrietal.,2017;Songetal.,2017;Salemetal.,2019),andthusthemodelsareconsideredtoDeepneural...
UnderstandingFailuresinOut-of-DistributionDetectionwithDeepGenerativeModelsLilyH.Zhang1MarkGoldstein1RajeshRanganath1Abstractdistribution,estimatedviaamodel(Bishop,1994).Deepgenerativemodels(DGMs)s...
StreamingBayesianDeepTensorFactorizationShikaiFang1ZhengWang1ZhimengPan1JiLiu23ShandianZhe1Abstractbeabletoestimateallkindsofcomplicated(e.g.,highlynonlinear)mappings.Themostrecentwork(Liuetal.,Des...
SpectralNormalisationforDeepReinforcementLearning:AnOptimisationPerspectiveFlorinGogianu12TudorBerariu3MihaelaRosca45ClaudiaClopath34LucianBusoniu2RazvanPascanu4AbstractFigure1:Optimisationrivalsal...
SimultaneousSimilarity-basedSelf-DistillationforDeepMetricLearningKarstenRoth12TimoMilbich2Bjo¨rnOmmer2JosephPaulCohen3MarzyehGhassemi41Abstractthatembeddingspacedimensionalitycanbeadriverforgener...
SignaturedDeepFictitiousPlayforMeanFieldGameswithCommonNoiseMingMin1RuimengHu12Abstractrealisticmodels,theproblemusuallylackstractabilityandneedsnumericalmethods.WithalargenumberofplayersExistingde...
Self-TuningforData-EfficientDeepLearningXimeiWang1JinghanGao1MingshengLong1JianminWang1Abstractmustbedonebyanexpertsuchasadoctorinmedicalapplications.Tomitigatetherequirementforlabeleddata,Deeplear...
SCC:anEfficientDeepReinforcementLearningAgentMasteringtheGameofStarCraftIIXiangjunWang1JunxiaoSong1PenghuiQi1PengPeng1ZhenkunTang1WeiZhang1WeiminLi1XiongjunPi1JujieHe1ChaoGao1HaitaoLong1QuanYuan1Ab...
ScalingPropertiesofDeepResidualNetworksAlain–SamCohen1RamaCont2AlainRossier21RenyuanXu2Abstractwhereh(kL)isthehiddenstateatlayerk=0,...,L,h(0L)=x∈Rdtheinput,h(LL)∈Rdtheoutput,σ:R→Risanon-Resid...
RevealingtheStructureofDeepNeuralNetworksviaConvexDualityTolgaErgen1MertPilanci1Abstractwheretwo-layerReLUnetworkswiththeminimumEu-clideannormsolutionandzerotrainingerrorareprovenWestudyregularized...
REPAINT:KnowledgeTransferinDeepReinforcementLearningYunzheTao1SahikaGenc1JonathanChung1TaoSun1SunilMallya1Abstractimproveperformanceonothertasks.AcceleratinglearningprocessesforcomplextasksTransfer...
WhittleNetworks:ADeepLikelihoodModelforTimeSeriesZhongjieYu1FabrizioVentola1KristianKersting12Abstract2015).Consider,e.g.,thestockpricechangesofindus-trialsectorsshowninFig.1(Left).Thedropinlate201...
Quasi-GlobalMomentum:AcceleratingDecentralizedDeepLearningonHeterogeneousDataTaoLin1SaiPraneethKarimireddy1SebastianU.Stich1MartinJaggi1Abstractiskeptlocally(nevertransmittedduringtraining).Decentr...
PracticalandPrivate(Deep)LearningWithoutSamplingorShufflingPeterKairouz1BrendanMcMahan1ShuangSong1OmThakkar1AbhraDeepThakurta1ZhengXu1Abstractinthecontextofdistributedsettingslikefederatedlearn-ing...
On-PolicyDeepReinforcementLearningfortheAverage-RewardCriterionYimingZhang1KeithW.Ross21AbstractHaarnojaetal.,2018)orinaqueuingscenario(Tadepalli&Ok,1994;Sutton&Barto,2018),thereisnonaturalsep-Wede...
RecoveringAESKeyswithaDeepColdBootAttackItamarZimerman1EliyaNachmani12LiorWolf1Abstractkeybyexploitingthememoryleakageandtheredundancyofthekeyexpansionfunctionusedbytheencryptionmethod.Coldbootatta...
Non-NegativeBregmanDivergenceMinimizationforDeepDirectDensityRatioEstimationMasahiroKato1TakeshiTeshima2Abstractthedensityratiowhenanomaly-freeunlabeledtestdataareavailable(Hidoetal.,2008).Densityr...
NeuralTransformationLearningforDeepAnomalyDetectionBeyondImagesChenQiu12TimoPfrommer1MariusKloft2StephanMandt3MajaRudolph1Abstractformationsareuseful,anditishardtodesignthesetrans-formationsmanuall...
MeasuringRobustnessinDeepLearningBasedCompressiveSensingMohammadZalbagiDarestani1AkshayS.Chaudhari2ReinhardHeckel13Abstractsiderthecompressivesensingproblemarisinginmagneticresonanceimaging(MRI),wh...
LeveragingSparseLinearLayersforDebuggableDeepNetworksEricWong1ShibaniSanturkar1AleksanderMa˛dry1Abstractmodel’sfailuremodesorevaluatecorrectiveinterventionswithoutin-depthproblem-specificstudies....