VectorQuantizedModelsforPlanningSherjilOzair12YazheLi1AliRazavi1IoannisAntonoglou1AäronvandenOord1OriolVinyals1AbstractDota2(OpenAIetal.,2019)andStarCraftII(Vinyalsetal.,2019),orrobotics(OpenAIeta...
TemporalPredictiveCodingForModel-BasedPlanningInLatentSpaceTungNguyen1RuiShu2TuanPham1HungBui1StefanoErmon2Abstractever,itiswellknownthatperformingRLdirectlyinthehigh-dimensionalobservationspaceiss...
SkillDiscoveryforExplorationandPlanningusingDeepSkillGraphsAkhilBagaria1JasonSenthil1GeorgeKonidaris1AbstractWeintroduceanewskill-discoveryalgorithmthatbuildsadiscretegraphrepresentationoflargecon-...
Self-ImprovedRetrosyntheticPlanningJunsuKim1SungsooAhn2HankookLee1JinwooShin1Abstract(a)ChemicalreactionRetrosyntheticPlanningisafundamentalproblem(b)RetrosyntheticPlanninginchemistryforfindingapat...
OptimizationPlanningfor3DConvNetsZhaofanQiu1TingYao1Chong-WahNgo2TaoMei1Abstractstance,anensembleofLGD-3Dnetworks(Qiuetal.,2019)achieves17.88%intermsofaverageerrorintrimmedvideoItisnottrivialtoopti...
LearningandPlanninginComplexActionSpacesThomasHubert1JulianSchrittwieser1IoannisAntonoglou1MohammadaminBarekatain1SimonSchmitt1DavidSilver1Abstractreal-worldproblems.Manyimportantreal-worldproblems...
LearningandPlanninginAverage-RewardMarkovDecisionProcessesYiWan1AbhishekNaik1RichardS.Sutton12Abstractwithit.Forlearningandcombinedmethods,bothcontrolandpredictionproblemscanbefurthersubdividedinto...
DynamicPlanningandLearningunderRecoveringRewardsDavidSimchi-Levi1ZeyuZheng2FengZhu1Abstractimmediatelydropsafteritispulled,andthengraduallyre-coversifthearmisnotpulledinthesubsequenttimeperiods.Mot...
DifferentiableSpatialPlanningusingTransformersDevendraSinghChaplot12DeepakPathak2JitendraMalik13Projectwebpage:https://devendrachaplot.github.io/projects/spatial-Planning-transformersAbstractFigure...
Task-OrientedActivePerceptionandPlanninginEnvironmentswithPartiallyKnownSemanticsMahsaGhasemi1ErdemArincBulgur2UfukTopcu2AbstractJointperceptionandPlanningposetwofundamentalchal-lenges.Thefirstchal...
SelectiveDyna-stylePlanningUnderLimitedModelCapacityMuhammadZaheer1SamuelSokota1ErinJ.Talvitie2MarthaWhite1Abstractdecisions.Incontrast,inmodel-basedreinforcementlearn-ing,theagentpossessesamodelof...
Retro:LearningRetrosyntheticPlanningwithNeuralGuidedASearchBinghongChen1ChengtaoLi2HanjunDai3LeSong14AbstractExistingmethodsroughlyfallintotwocategories,eithertemplate-basedortemplate-free.Eachchem...
PlanningtoExploreviaSelf-SupervisedWorldModelsRamananSekar1OlehRybkin1KostasDaniilidis1PieterAbbeel2DanijarHafner34DeepakPathak56AbstractModelLearningReinforcementlearningallowssolvingcomplexTaskAt...
PackIt:AVirtualEnvironmentforGeometricPlanningAkitGoyal1JiaDeng1AbstractlikerobotstopossesstheabilityofgeometricPlanningsothattheycanoperateinunconstrainedhumanenvironmentsTheabilitytojointlyunders...
OnValidationandPlanningofAnOptimalDecisionRulewithApplicationinHealthcareStudiesHengruiCai1WenbinLu1RuiSong1Abstractsionforallindividuals.Anumberofmethodshavebeendevelopedforestimatingoptimaldecisi...
LearningPortableRepresentationsforHigh-LevelPlanningStevenJames1BenjaminRosman1GeorgeKonidaris2AbstractOneapproachistobuildastateabstractionoftheenviron-mentthatsupportsPlanning.Suchrepresentations...
HallucinativeTopologicalMemoryforZero-ShotVisualPlanningKaraLiu1ThanardKurutach1ChristineTung1PieterAbbeel1AvivTamar12Abstracttemdynamicsarenotknown,andonlyadatasetofstatetransitionsisavailable.Inp...
FlexibleandEfficientLong-RangePlanningThroughCuriousExplorationAidanCurtis1MinjianXin2DilipArumugam3KevinFeigelis3DanielYamins3Abstract1.IntroductionIdentifyingalgorithmsthatflexiblyandefficientlyM...
Scale-freeadaptivePlanningfordeterministicdynamics&discountedrewardsPeterL.Bartlett1VictorGabillon2JenniferHealey3MichalValko4Abstractsetting.Ouralgorithmimplementsascale-freefunctionop-timizations...