Reinforcement learning (RL) has shown great success in increasingly complex single-agent environments and two-player turn-based games. However, the real world contains multiple agents, each learning and acting independently to cooperate and compete with other agents. We used a tournament-style evaluation to demonstrate that an agent can achieve human-level performance in a three-dimensional multiplayer first-person video game, Quake III Arena in Capture the Flag mode, using only pixels and game points scored as input. We used a two-tier optimization process in which a population of independent RL agents are trained concurrently from thousands of parallel matches on randomly generated environments. Each agent learns its own internal reward signal and rich representation of the world. These results indicate the great potential of multiagent reinforcement learning for artificial intelligence research.


    Access

    Download


    Export, share and cite



    Title :

    Human-level performance in 3D multiplayer games with population-based reinforcement learning


    Contributors:
    Jaderberg, M (author) / Czarnecki, WM (author) / Dunning, I (author) / Marris, L (author) / Lever, G (author) / Castaneda, AG (author) / Beattie, C (author) / Rabinowitz, NC (author) / Morcos, AS (author) / Ruderman, A (author)

    Publication date :

    2019-05-31


    Remarks:

    Science , 364 (6443) pp. 859-865. (2019)


    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English



    Classification :

    DDC:    629



    Suboptimal Approaches to Multiplayer Pursuit-Evasion Differential Games

    Ge, Jianhua / Tang, Liang / Reimann, Johan et al. | AIAA | 2006



    AIAA-2006-6786 Suboptimal Approaches to Multiplayer Pursuit-Evasion Differential Games

    Ge, J. / Tang, L. / Reimann, J. et al. | British Library Conference Proceedings | 2006


    An Optimistic Obsolescence-based Approach to Event Synchronization for Massively Multiplayer Online Games

    Ferretti, S. / Roccetti, M. / Palazzi, C. E. | British Library Online Contents | 2007