AFullyDifferentiableBeamSearchDecoderRonanCollobert1AwniHannun1GabrielSynnaeve1Abstractownmistakessinceitusestheground-truthtargetasguid-ance.AtinferencethetargetisunavailableandabeamWeintroduceane...
ABetterk-means++AlgorithmviaLocalSearchSilvioLattanzi1ChristianSohler1AbstractThek-means++seedingalgorithm(Arthur&Vassilvitskii,2007)isasimplewaytoimproveLloyd’salgorithm.TheInthispaper,wedevelopa...
PIPPS:FlexibleModel-BasedPolicySearchRobusttotheCurseofChaosPaavoParmas1CarlEdwardRasmussen2JanPeters34KenjiDoya1AbstractVelocityPreviously,theexplodinggradientproblemhasPositionPositionbeenexplain...
Path-LevelNetworkTransformationforEfficientArchitectureSearchHanCai1JiachengYang1WeinanZhang1SongHan2YongYu1Abstractthisprocesstypicallyrequiresyearsofextensiveinvestiga-tionbyhumanexperts,whichisn...
ImprovednearestneighborSearchusingauxiliaryinformationandpriorityfunctionsOmidKeivani1KaushikSinha1AbstractThenaivelineartimesolution,thatscansthrougheachdatapointxi∈S,oftenbecomesimpracticalforla...
Feedback-BasedTreeSearchforReinforcementLearningDanielR.Jiang1EmmanuelEkwedike23HanLiu24Abstractleaf-nodeevaluators(eitherapolicyfunction(Chaslotetal.,2006)rollout,avaluefunctionevaluation(Campbell...
EfficientNeuralArchitectureSearchviaParameterSharingHieuPham12MelodyY.Guan3BarretZoph1QuocV.Le1JeffDean1Abstractconsuming,e.g.Zophetal.(2018)use450GPUsfor3-4days(i.e.32,400-43,200GPUhours).Meanwhil...
EfficientGradient-FreeVariationalInferenceusingPolicySearchOlegArenz1MingjunZhong2GerhardNeumann13Abstractuseitforinference,acommonapproachistouseVaria-tionalInference(VI)toapproximatethetargetdist...
NeuralOptimizerSearchwithReinforcementLearningIrwanBello1BarretZoph1VijayVasudevan1QuocV.Le1AbstractFigure1.AnoverviewofNeuralOptimizerSearch.Wepresentanapproachtoautomatetheprocessrentnetworkcontr...
Max-valueEntropySearchforEfficientBayesianOptimizationZiWang1StefanieJegelka1AbstractAmongthemostpopularonesrangetheGaussianprocessupperconfidencebound(GP-UCB)(Auer,2002;SrinivasEntropySearch(ES)an...
Fastk-NearestNeighbourSearchviaPrioritizedDCI11KeLiJitendraMalikAbstractcurseofdimensionality,whichdescribesthephenomenonofquerytimecomplexitydependingexponentiallyondi-Mostexactmethodsfork-nearest...
EfficientNonmyopicActiveSearchShaliJiang1GustavoMalkomes1GeoffConverse2AlyssaShofner3BenjaminMoseley1RomanGarnett1Abstractingamodeltohavehighgeneralizationperformancewithfewtrainingexamples.Here,we...
Data-EfficientPolicyEvaluationThroughBehaviorPolicySearchJosiahP.Hanna1PhilipS.Thomas23PeterStone1ScottNiekum1AbstractMethodsthatevaluateπewhileselectingactionsaccordingtoπearetermedon-policy.Pre...