Corpus ID: 25026734

Value-Decomposition Networks For Cooperative Multi-Agent Learning

@article{Sunehag2018ValueDecompositionNF,
  title={Value-Decomposition Networks For Cooperative Multi-Agent Learning},
  author={Peter Sunehag and G. Lever and A. Gruslys and W. Czarnecki and V. Zambaldi and Max Jaderberg and Marc Lanctot and Nicolas Sonnerat and Joel Z. Leibo and K. Tuyls and T. Graepel},
  journal={ArXiv},
  year={2018},
  volume={abs/1706.05296}
}
  • Peter Sunehag, G. Lever, +8 authors T. Graepel
  • Published 2018
  • Computer Science
  • ArXiv
  • We study the problem of cooperative multi-agent reinforcement learning with a single joint reward signal. This class of learning problems is difficult because of the often large combined action and observation spaces. In the fully centralized and decentralized approaches, we find the problem of spurious rewards and a phenomenon we call the "lazy agent" problem, which arises due to partial observability. We address these problems by training individual agents with a novel value decomposition… CONTINUE READING
    221 Citations

    Figures and Topics from this paper

    Towards Understanding Linear Value Decomposition in Cooperative Multi-Agent Q-Learning
    • 4
    • Highly Influenced
    • PDF
    Q-value Path Decomposition for Deep Multiagent Reinforcement Learning
    • 6
    • PDF
    Graph Convolutional Value Decomposition in Multi-Agent Reinforcement Learning
    • Highly Influenced
    • PDF
    Is Independent Learning All You Need in the StarCraft Multi-Agent Challenge?
    • PDF
    QOPT: Optimistic Value Function Decentralization for Cooperative Multi-Agent Reinforcement Learning
    • 2
    • Highly Influenced
    Learning Transferable Cooperative Behavior in Multi-Agent Teams
    • 21
    • PDF

    References

    SHOWING 1-10 OF 49 REFERENCES
    The Dynamics of Reinforcement Learning in Cooperative Multiagent Systems
    • 1,094
    • Highly Influential
    • PDF
    Cooperative Multi-Agent Learning: The State of the Art
    • 1,138
    • PDF
    Q-Decomposition for Reinforcement Learning Agents
    • 126
    • PDF
    Learning to Communicate with Deep Multi-Agent Reinforcement Learning
    • 619
    • PDF
    Coordinated Reinforcement Learning
    • 332
    • PDF
    Modeling Difference Rewards for Multiagent Learning (Extended Abstract)
    • 6
    A Comprehensive Survey of Multiagent Reinforcement Learning
    • 1,229
    Modeling difference rewards for multiagent learning
    • 19
    • PDF
    Analyzing and visualizing multiagent rewards in dynamic and stochastic domains
    • 99
    • PDF
    Combining reward shaping and hierarchies for scaling to large multiagent systems
    • 12
    • PDF