CIFS:ImprovingAdversarialRobustnessofCNNsviaChannel-wiseImportance-basedFeatureSelectionHanshuYan1JingfengZhang2GangNiu2JiashiFeng1VincentY.F.Tan13MasashiSugiyama24Abstracttonomousvehicles(Grigores...
AccelerateCNNsfromThreeDimensions:AComprehensivePruningFrameworkWenxiaoWang12MinghaoChen1ShuaiZhao1LongChen34JinmingHu1HaifengLiu1DengCai1XiaofeiHe1WeiLiu2Abstract57Mostneuralnetworkpruningmethods,...
InterpolationbetweenResidualandNon-ResidualNetworksZonghanYang1YangLiu1ChenglongBao2ZuoqiangShi3Abstracteringadifferentialequationthatidentifiestheruleoftheobserveddatabasedonthestandardblockofexis...
UnderstandingGeometryofEncoder-DecoderCNNsJongChulYe12WoonKyoungSung2Abstractpower,generalizationcapability,andoptimizationland-scapeofDNNshavebecomeanintellectualchallengeforEncoder-decodernetwork...
TrainingCNNswithSelectiveAllocationofChannelsJongheonJeong1JinwooShin123AbstractChenetal.,2018)andanytime/adaptivenetworks(Figurnovetal.,2017;Bolukbasietal.,2017;Huangetal.,2018a).Recentprogressind...
OptimizationLandscapeandExpressivityofDeepCNNsQuynhNguyen1MatthiasHein2AbstractTable1.Themaximumwidthofalllayersinseveralstate-of-the-artCNNarchitecturescomparedwiththesizeofImageNetdatasetWeanalyz...