Learn-to-Share:AHardware-friendlyTransferLearningFrameworkExploitingComputationandParameterSharingChengFu1HanxianHuang1XinyunChen2YuandongTian3JishenZhao1AbstractFigure1:RoadmapofGLUEscorev.s.total...
ExploitingStructuredDataForLearningContagiousDiseasesUnderIncompleteTestingMaggieMakar1LaurenWest2DavidHooper2EricHorvitz3EricaShenoy2JohnGuttag1Abstractiidassumptioncentraltomostmachinelearningmet...
ExploitingSharedRepresentationsforPersonalizedFederatedLearningLiamCollins1HamedHassani2AryanMokhtari1SanjayShakkottai1Abstracteffectivemodelsforeachclientbyleveragingthelocalcom-putationalpower,me...
PuzzleMix:ExploitingSaliencyandLocalStatisticsforOptimalMixupJang-HyunKim12000Choo12HyunOhSong12Wo0n0h1o002003Abstract004Input1InputMixupPuzzleMix(zonly)Whiledeepneuralnetworksachievegreatper-005fo...
InducingandExploitingActivationSparsityforFastNeuralNetworkInferenceMarkKurtz1JustinKopinsky1RatiGelashvili1AlexanderMatveev1JohnCarr1MichaelGoin1WilliamLeiserson1SageMoore1BillNell1NirShavit1DanAl...
EnhancingSimpleModelsbyExploitingWhatTheyAlreadyKnowAmitDhurandhar1KarthikeyanShanmugam1RonnyLuss1Abstract1.IntroductionTherehasbeenrecentinterestinimprovingper-Simplemodelssuchasdecisiontreesorrul...
GEOMetrics:ExploitingGeometricStructureforGraph-EncodedObjectsEdwardJ.Smith1ScottFujimoto12AdrianaRomero31DavidMeger1Abstract(a)Voxels(b)Pointcloud(262144units)(30000points)Meshmodelsareapromisinga...
ExploitingWorkerCorrelationforLabelAggregationinCrowdsourcingYuanLi1BenjaminI.P.Rubinstein1TrevorCohn1Abstractequalvotestowardsconsensus.Numerousprobabilisticmodelshaveemergedthatparameteriseworker...
ExploitingStructureofUncertaintyforEfficientMatroidSemi-BanditsPierrePerrault12VianneyPerchet23MichalValko14Abstracttimeandspacecomplexity)oftheircombinatorialgeneral-ization,inwhichtheagentselects...
ExploitingthePotentialofStandardConvolutionalAutoencodersforImageRestorationbyEvolutionarySearchMasanoriSuganuma12MeteOzay2TakayukiOkatani21Abstractthelimitationofthesehand-designedorsimplepriors,a...
ExploitingStrongConvexityfromDatawithPrimal-DualFirst-OrderAlgorithmsJialeiWang1LinXiao2Abstractfunctionisφi(z)=(1/2)(z−bi)2,andwegetridgere-gressionwith2andtheelasticnetwithWeconsiderempiricalri...