UnifyingVision-and-LanguageTasksviaTextGenerationJaeminCho1JieLeiHaoTanMohitBansalUNCChapelHill{jmincho,jielei,haotan,mbansal}@cs.unc.eduAbstract21Existingmethodsforvision-and-languagelearn-ingtypi...
TreeswithAttentionforSetPredictionTasksRoyHirsch1RanGilad-Bachrach2AbstractTree-basedmodels,suchasDecisionTree(DT),RandomForest(RF)andGradientBoostingDecisionTree(GBT)Inmanymachinelearningapplicati...
SolvingChallengingDexterousManipulationTasksWithTrajectoryOptimisationandReinforcementLearningHenryCharlesworth1GiovanniMontana1Abstractthehumanhand—capableofTasksrangingfromcomplexgraspingtowriti...
LocallyPersistentExplorationinContinuousControlTaskswithSparseRewardsSusanAmin12MaziarGomrokchi12HosseinAboutalebi34HarshSajita12DoinaPrecup12Abstractcallforacleverexplorationstrategythatexposesthe...
Few-shotConformalPredictionwithAuxiliaryTasksAdamFisch1TalSchuster1TommiJaakkola1ReginaBarzilay1Abstractforcreatingconfidence-basedsetpredictions(Vovketal.,2005).Concretely,supposewehavebeengivenne...
LearningCompoundTaskswithoutTask-specificKnowledgeviaImitationandSelf-supervisedLearningSang-HyunLee12Seung-WooSeo1AbstractCompoundTasksub-task3Mostreal-worldTasksarecompoundTasksthatContinuousLate...
TransferLearningforRelatedReinforcementLearningTasksviaImage-to-ImageTranslationShaniGamrian1YoavGoldberg12Abstractprocessofanewtaskhastobeperformedfromscratchevenforarelatedone.Recentworkshavetrie...
LeveragingLow-RankRelationsBetweenSurrogateTasksinStructuredPredictionGiuliaLuise1DimitrisStamos1MassimilianoPontil12CarloCiliberto13AbstractAmongthemostwell-establishedstrategiesforstructuredpredi...
LearningbyPlaying–SolvingSparseRewardTasksfromScratchMartinRiedmiller1RolandHafner1ThomasLampe1MichaelNeunert1JonasDegrave1TomVandeWiele1VolodymyrMnih1NicolasHeess1TobiasSpringenberg1Abstractsimul...
Multi-taskLearningwithLabeledandUnlabeledTasksAnastasiaPentina1ChristophH.Lampert1Abstractisavailable.Inpractice,itishighlydesirabletobeabletohandlethissituationforproblemswithaverylargenumberInmul...