OPtions as REsponses Grounding behavioural hierarchies in multi-agent reinforcement learning

OPtionsasREsponses:GroundingBehaviouralHierarchiesinMulti-AgentReinforcementLearningAlexanderSashaVezhnevets1YuhuaiTonyWu12MariaEckstein31Re´miLeblond1JoelZ.Leibo1Abstractproblemofbuildingagentswi...