OnlineOptimizationinGamesviaControlTheory:ConnectingRegret,PassivityandPoincare´RecurrenceYunKuenCheung1GeorgiosPiliouras2Abstracttheoryandonlineoptimization.Wepresentanovelcontrol-theoreticunders...
OnlineLearninginUnknownMarkovGamesYiTian1YuanhaoWang2TianchengYu1SuvritSra1Abstractcontrolboth/allplayersandaimtominimizethenumberofepisodesrequiredtofindagoodpolicy;and(2)theonlineWestudyonlinelea...
LearningWhilePlayinginMean-FieldGames:ConvergenceandOptimalityQiaominXie1ZhuoranYang2ZhaoranWang3AndreeaMinca1Abstractfromthescalabilityissue.Specifically,inamulti-agentsystem,eachagentinteractswit...
LearninginNonzero-SumStochasticGameswithPotentialsDavidMguni1YutongWu2YaliDu3YaodongYang13ZiyiWang2MinneLi3YingWen4JoelJennings1JunWang3Abstractautonomousvehiclesseekingtoarriveattheirindividualdes...
Infinite-DimensionalOptimizationforZero-SumGamesviaVariationalTransportLewisLiu1YufengZhang2ZhuoranYang3RezaBabanezhad4ZhaoranWang2Abstract(NNs)),whichisofindependentinterest.Gameoptimizationhasbee...
Follow-the-Regularized-LeaderRoutestoChaosinRoutingGamesJakubBielawski1ThiparatChotibut2FryderykFalniowski1GrzegorzKosiorowski1MichałMisiurewicz3GeorgiosPiliouras4Abstractwithachoicebetweentwostra...
DiscretizationDriftinTwo-PlayerGamesMihaelaRosca12YanWu1BenoitDherin3DavidG.T.Barrett1AbstractoftwoplayerGamesbyfindingcontinuoussystemswhichbettermatchthegradientdescentupdatesusedinpractice.Gradi...
AdversarialPolicyLearninginTwo-playerCompetitiveGamesWenboGuo1XianWu1SuiHuang2XinyuXing1Abstract2020),wearguethatattacksdevelopedunderthisassump-tionarenotpractical.Forexample,givenamasteragentInat...
StochasticRegretMinimizationinExtensive-FormGamesGabrieleFarina1ChristianKroer2TuomasSandholm1345AbstractTypically,EFGmodelsareoperationalizedbycomputingeitheraNashequilibriumofthegame,oranapproxim...
StochasticHamiltonianGradientMethodsforSmoothGamesNicolasLoizou1HugoBerard12AlexiaJolicoeur-Martineau1PascalVincent†12SimonLacoste-Julien†1IoannisMitliagkas†1Abstractforeveryx1∈Rd1andx2∈Rd2.We...
Low-VarianceandZero-VarianceBaselinesforExtensive-FormGamesTrevorDavis1†MartinSchmid2MichaelBowling21Abstractetal.,2015),andtobeathumanprofessionalsinanother(Moravcˇíketal.,2017;Brown&Sandholm,2...
LearningQuadraticGamesonNetworksYanLeng1XiaowenDong2JunfengWu3AlexPantland4AbstractFacebook,onlytofindoutthatanypairofFacebookuserscanactuallybeconnectedviaapproximatelythreeandahalfIndividuals,oro...
InvariantRiskMinimizationGamesKartikAhuja1KarthikeyanShanmugam1KushR.Varshney1AmitDhurandhar1Abstractwereindeserts.TheCNNpickedupthespuriouscorrela-tion,i.e.,itassociatedgreenpastureswithcowsandfai...
ImplicitLearningDynamicsinStackelbergGames:EquilibriaCharacterization,ConvergenceAnalysis,andEmpiricalStudyTannerFiez1BenjaminChasnov1LillianRatliff1Abstractandhyperparameteroptimization(Maclaurine...
Gradient-freeOnlineLearninginGameswithDelayedRewardsAmélieHéliou1PanayotisMertikopoulos21ZhengyuanZhou3AbstractSimilarissuesalsoariseinoperationsresearch,onlinemachinelearning,andotherfieldswhere...
Open-endedLearninginSymmetricZero-sumGamesDavidBalduzzi1MartaGarnelo1YoramBachrach1WojciechM.Czarnecki1JulienPerolat1MaxJaderberg1ThoreGraepel1Abstractofwhattesttotake,orwhatobjectivetooptimize,isn...
TheMechanicsofn-PlayerDifferentiableGamesDavidBalduzzi1Se´bastienRacanie`re1JamesMartens1JakobFoerster2KarlTuyls1ThoreGraepel1Abstractoptimization(Pfau&Vinyals,2016),syntheticgradients(Jaderberget...
InvestigatingHumanPriorsforPlayingVideoGamesRachitDubey1PulkitAgrawal1DeepakPathak1ThomasL.Griffiths1AlexeiA.Efros1AbstractFigure1.Motivatingexample.(a)Asimpleplatformergame.(b)Thesamegamemodifiedb...
CompilingCombinatorialPredictionGamesFredericKoriche1AbstractConceptually,anonlinecombinatorialoptimizationproblemInonlineoptimization,thegoalistoiterativelycanbecastasarepeatedpredictiongamebetwee...
CanDeepReinforcementLearningSolveErdos-Selfridge-SpencerGames?MaithraRaghu12AlexIrpan1JacobAndreas3RobertKleinberg2QuocLe1JonKleinberg2Abstractbehaviorisdifficult.Optimalbehaviorintheseenviron-ment...