GeneralizationGuaranteesforNeuralArchitectureSearchwithTrain-ValidationSplitSametOymak1MingchenLi2MahdiSoltanolkotabi3Abstractportantfordeeplearningapplicationswheretherearemanypossibilitiesforchoo...
KNAS:GreenNeuralArchitectureSearchJingjingXu1LiangZhao2JunyangLin3RundongGao1XuSun12HongxiaYang3Abstracttecture.Theoptimizationmethoddictateshowtoexplorethesearchspace.Architectureevaluationisrespo...
Few-shotNeuralArchitectureSearchYiyangZhao1LinnanWang2YuandongTian3RodrigoFonseca2TianGuo1AbstractOnearchitectureEfficientevaluationofanetworkarchitecture?:entire?drawnfromalargesearchspaceremain...
Expressive1-LipschitzNeuralNetworksforRobustMultipleGraphLearningagainstAdversarialAttacksXinZhao1ZeruZhang1ZijieZhang1LingfeiWu2JiayinJin1YangZhou1RuomingJin3DejingDou45DaYan6Abstractment)(Zhang&T...
ExponentiallyManyLocalMinimainQuantumNeuralNetworksXuchenYou12XiaodiWu12Abstractso-calledvariationalquantummethod(e.g.,Peruzzoetal.(2014)),arethemajorcandidatesofapplicationsthatcanbeQuantumNeuralN...
EnhancingRobustnessofNeuralNetworksthroughFourierStabilizationNetanelRaviv1AidanKelley1MichaelGuo1YevgenyVorobeychik1Abstractevasionattacks(Grosseetal.,2016;Li&Vorobeychik,2018;Laskovetal.,2014;Xue...
EfficientStatisticalTests:ANeuralTangentKernelApproachShengJia12EhsanNezhadarya3YuhuaiWu12JimmyBa12Abstractformaximizingthetestpower(Liuetal.,2020;Gaoetal.,2020).Thesemethodshavebeenshowntoempirica...
E(n)EquivariantGraphNeuralNetworksVictorGarciaSatorras1EmielHoogeboom1MaxWelling1AbstractThispaperintroducesanewmodeltolearngraphNeuralnetworksequivarianttorotations,transla-tions,reflectionsandper...
FL-NTK:ANeuralTangentKernel-basedFrameworkforFederatedLearningAnalysisBaiheHuang1XiaoxiaoLi2ZhaoSong3XingYang4Abstractfrompersonaldata.Thesepersonaldataaresubjecttoreg-ulationssuchasCaliforniaConsu...
CyclicallyEquivariantNeuralDecodersforCyclicCodesXiangyuChen1MinYe1Abstractcodes,polarcodes,LDPCcodesandReed-Muller(RM)codes.Neuraldecoderswereintroducedasageneraliza-tionoftheclassicBeliefPropagat...
DynamicGameTheoreticNeuralOptimizerGuan-HorngLiu12TianrongChen3EvangelosA.Theodorou12AbstractFigure1.DynamicgameperspectiveofgenericDNNtrainingpro-cess,wherewetreatlayermodulesasplayersinadynamicga...
ConditionalTemporalNeuralProcesseswithCovarianceLossBoseonYoo1JiwooLee2JanghoonJu2SeijunChung1SoyeonKim1JaesikChoi13Abstractducedin(Garneloetal.,2018a),whicharethenoveleffortsthatimplementsuchGPswi...
CATE:Computation-awareNeuralArchitectureEncodingwithTransformersShenYan1KaiqiangSong23FeiLiu2MiZhang1Abstract2020)ordesigningefficientarchitecturesearchandevalu-ationmethods(Luoetal.,2018b;Shietal....
BackdoorScanningforDeepNeuralNetworksthroughK-ArmOptimizationGuangyuShen1YingqiLiu1GuanhongTao1ShengweiAn1QiulingXu1SiyuanCheng1ShiqingMa2XiangyuZhang1Abstractdoor)attackexploitstheuninterpretabili...
AdversarialRobustnessGuaranteesforRandomDeepNeuralNetworksGiacomoDePalma123BobakT.Kiani34SethLloyd23Abstractanextremelysmallperturbationcanchangethelabelofacorrectlyclassifedimage.Thispropertyposes...
AdaXpert:AdaptingNeuralArchitectureforGrowingDataShuaichengNiu12JiaxiangWu3GuanghuiXu1YifanZhang4YongGuo1PeilinZhao3PengWang5MingkuiTan16Abstract80networkadjustmentInreal-worldapplications,dataofte...
AUnifiedLotteryTicketHypothesisforGraphNeuralNetworksTianlongChen1YongduoSui2XuxiChen1AstonZhang3ZhangyangWang1Abstract0.0059.0486.580.0083.2294.5097.75WithgraphsrapidlygrowinginsizeanddeeperGNNSpa...
AReceptorSkeletonforCapsuleNeuralNetworksJintaiChen1HongyunYu1ChengdeQian2DannyZ.Chen3JianWu4AbstractSeniorRepresentationsInpreviousCapsuleNeuralNetworks(CapsNets),routingroutingalgorithmsoftenperf...
AProbabilisticApproachtoNeuralNetworkPruningXinQian1DiegoKlabjan1AbstractcalledtheLotteryTicketHypothesis(LTH).Ramanujanetal.(2020)furtherconjecturethat,asufficientlyover-Neuralnetworkpruningtechni...