TowardsUnderstandingtheRegularizationofAdversarialRobustnessonNeuralNetworksYuxinWen12ShuaiLi1KuiJia12Abstract(Szegedyetal.,2014).Moreintriguingly,theseadversarialexamplesaregeneratedbyaddingimperc...
TowardsUnderstandingtheDynamicsoftheFirst-OrderAdversariesZhunDeng1HangfengHe2JiaoyangHuang3WeijieJ.Su2Abstractattackssuchasinputde-noising(Guoetal.,2017),random-izedsmoothing(Ilyasetal.,2019),grad...
TaskUnderstandingfromConfusingMulti-taskDataXinSu12YizhouJiang1ShangqiGuo13FengChen132AbstractTaskAnnotationLabelAnnotationBeyondmachinelearning’ssuccessinthespecificTaskTAasnkno1tationSampleAnnot...
InverseActiveSensing:ModelingandUnderstandingTimelyDecision-MakingDanielJarrett1MihaelavanderSchaar12Abstractconductsacquisitionsforinformation(i.e.hypothesistests,diagnosticprocedures,andcandidate...
ExplainableandDiscourseTopic-awareNeuralLanguageUnderstandingYatinChaudhary12HinrichSchütze2PankajGupta1Abstracttributionoverwordsinvocabulary.Beyondadocumentrepresentation,topicmodelsalsoofferint...
UnderstandingtheOriginsofBiasinWordEmbeddingsMarc-EtienneBrunet12ColleenAlkalay-Houlihan1AshtonAnderson12RichardZemel12Abstract2013a)andGloVe(Penningtonetal.,2014)acquirestereo-typicalhumanbiasesfr...
UnderstandingtheImpactofEntropyonPolicyOptimizationZafaraliAhmed12NicolasLeRoux13MohammadNorouzi3DaleSchuurmans34Abstractlis,2000;Greensmithetal.,2004;Schulmanetal.,2015b;Tuckeretal.,2018).Entropyr...
UnderstandingPriorsinBayesianNeuralNetworksattheUnitLevelMariiaVladimirova12JakobVerbeek1PabloMesejo3JulyanArbel1AbstractworksonBayesianneuralnetworks(Neal,1992;MacKay,1992).Thereisalargevarietyofa...
UnderstandingMCMCDynamicsasFlowsontheWassersteinSpaceChangLiu1JingweiZhuo1JunZhu1AbstractdistributionminimizestheKLdivergencetothetargetdistri-bution.TheyfullyexploittheapproximationabilityofasetIt...
UnderstandingGeometryofEncoder-DecoderCNNsJongChulYe12WoonKyoungSung2Abstractpower,generalizationcapability,andoptimizationland-scapeofDNNshavebecomeanintellectualchallengeforEncoder-decodernetwork...
UnderstandingImpactsofHigh-OrderLossApproximationsandFeaturesinDeepLearningInterpretationSahilSingla1EricWallace1ShiFeng1SoheilFeizi1Abstracthindmodelpredictions?Acommoninterpretationapproachistoid...
UnderstandingandUtilizingDeepNeuralNetworksTrainedwithNoisyLabelsPengfeiChen12BenbenLiao2GuangyongChen2ShengyuZhang12AbstractbasedonwhichwecandesignspecificmethodstotrainDNNsrobustlyinpracticalappl...
UnderstandingandcorrectingpathologiesinthetrainingoflearnedoptimizersLukeMetz1NiruMaheswaranathan1JeremyNixon1C.DanielFreeman1JaschaSohl-Dickstein1Abstractperparametersearch.Ontheotherhand,acomplem...
UnderstandingandControllingMemoryinRecurrentNeuralNetworksDoronHaviv12AlexnaderRivkind234OmriBarak23Abstract1.IntroductionTobeeffectiveinsequentialdataprocessing,Re-RecurrentNeuralNetworks(RNN)aret...
UnderstandingandAcceleratingParticle-BasedVariationalInferenceChangLiu1JingweiZhuo1PengyuCheng2RuiyiZhang2JunZhu1LawrenceCarin2Abstractmethods(Geman&Geman,1987;Nealetal.,2011;Welling&Teh,2011;Dinge...
TowardsUnderstandingKnowledgeDistillationMaryPhuong1ChristophH.Lampert1Abstractdistillation-basedtraininghasbeenconfirmedseveraltimes:theoptimizationstepisgenerallymorewell-behavedthanKnowledgedist...
TowardsaDeepandUnifiedUnderstandingofDeepNeuralModelsinNLP1ChaoyuGuan2XitingWang2QuanshiZhang1RunjinChen1DiHe3XingXie2Abstract(a)Gradient-basedmethod(b)OursWedefineaunifiedinformation-basedmeasureF...
TowardsUnderstandingtheImportanceofNoiseinTrainingNeuralNetworksMoZhou1TianyiLiu2YanLi2DachaoLin1EnluZhou2TuoZhao2AbstractSimplefirstorderalgorithmssuchasStochasticGradientDescent(SGD)anditsvariant...
Shallow-DeepNetworks:UnderstandingandMitigatingNetworkOverthinkingYigitcanKaya1SanghyunHong1TudorDumitras,1Abstractperformevencomplextasks,suchasdrivingorplayingthepiano,withlittlementaleffort(Kahn...
AnalogiesExplained:TowardsUnderstandingWordEmbeddingsCarlAllen1TimothyHospedales1Abstractofwordwi.Thisenablesanalogicalquestionssuchas“manistokingaswomanisto..?”tobesolvedbyvectoraddi-Wordembeddi...