TrajectoryDiversityforZero-ShotCoordinationAndreiLupu1BrandonCui2HengyuanHu2JakobFoerster2AbstractFigure1:ThecorridorCoordinationtaskadmitsthreeopti-malSPpolicies,onlyoneofwhichissuitableforZSC.Wes...
Few-shotLanguageCoordinationbyModelingTheoryofMindHaoZhu1GrahamNeubig1YonatanBisk1Abstractprocess,withawidevarietyofworksexaminingcommuni-cationbetweenagentsviaeithercompletelyartificialemer-Nomani...
ContinuousCoordinationAsaRealisticScenarioforLifelongLearningHadiNekoei1AkileshBadrinaaraayanan12AaronCourville123SarathChandar143Abstract1.IntroductionCurrentdeepreinforcementlearning(RL)algo-Deep...
DeepCoordinationGraphsWendelinBo¨hmer1VitalyKurin1ShimonWhiteson1AbstractConsequently,thejointvaluefunctioncanbeefficientlymaximizedifeachagentsimplyselectstheactionthatmax-Thispaperintroducesthed...
“Other-Play”forZero-ShotCoordinationHengyuanHu1AdamLerer1AlexPeysakhovich1JakobFoerster1AbstractOursettingisapartiallyobservedcooperativeMarkovgame(MG)whichiscommonlyknownamongbothagents.TheWecon...
LearningtoCoordinatewithCoordinationGraphsinRepeatedSingle-StageMulti-AgentDecisionProblemsEugenioBargiacchi1TimothyVerstraeten1DiederikM.Roijers12AnnNowe´1HadovanHasselt3Abstractoverlapping)subse...