ScalingUpVisualandVision-LanguageRepresentationLearningWithNoisyTextSUpervisionChaoJia1YinfeiYang1YeXia1Yi-TingChen1ZaranaParekh1HieUpham1QuocV.Le1YunhsuanSung1ZhenLi1TomDuerig1Abstract1.Introducti...
NewtonMethodoverNetworksisFastUptotheStatisticalPrecisionAmirDaneshmand1GesualdoScutari1PavelDvurechensky23AlexanderGasnikov43Abstractwhere:Rd×Z→Risthelossfunction,assumedtobe(strongly)convexinx,...
UporDown?AdaptiveRoundingforPost-TrainingQuantizationMarkusNagel1RanaAliAmjad1MartvanBaalen1ChristosLouizos1TijmenBlankevoort1Abstractcations,andevendedicatedlow-powerhardware.Whenquantizingneuraln...
Small-GAN:SpeedingUpGANTrainingusingCore-SetsSamarthSinha1HanZhang2AnirudhGoyal3YoshuaBengio3HugoLarochelle23AugustusOdena2AbstractRecently,Brocketal.(2018)substantiallyimprovedtheresultsofZhangeta...
ScalingUpHybridProbabilisticInferencewithLogicalandArithmeticConstraintsviaMessagePassingZheZeng1PaoloMorettin2FanqiYan3AntonioVergari1GuyVandenBroeck1Abstractmodelforthistaskhastodealwithcontinuou...
FastandThree-rious:SpeedingUpWeakSUpervisionwithTripletMethodsDanielY.Fu1MayeeF.Chen1FredericSala1SarahM.Hooper2KayvonFatahalian1ChristopherRe´1Abstract2012);itdrivesapplicationsusedbybillionsofpe...
TradingRedundancyforCommunication:SpeedingUpDistributedSGDforNon-convexOptimizationFarzinHaddadpour1MohammadMahdiKamani2MehrdadMahdavi1ViveckR.Cadambe1Abstractchinelearningmodelssuchasdeepneuralnet...
ScalingUpOrdinalEmbedding:ALandmarkApproachJesseAnderton1JavedAslam1Abstractpositionsanddistancesarenotreadilyavailable.Itwasorig-inallyusedwithtripletssourcedfromhumancomparisonsOrdinalEmbeddingis...
LearningtoSpeedUpStructuredOutputPredictionXingyuanPan1VivekSrikumar1AbstractInthispaper,wefocusonthecomputationalcostofin-ference.WearguethatnaturallyoccurringproblemshavePredictingstructuredoutpu...
ScalingUpSparseSUpportVectorMachinesbySimultaneousFeatureandSampleReductionWeizhongZhang12BinHong13WeiLiu2JiepingYe3DengCai1XiaofeiHe1JieWang3Abstracttionandvariableselectionby1-normpenalty.Thelast...