Corpus ID: 67751744

Communication Topologies Between Learning Agents in Deep Reinforcement Learning

@article{Adjodah2019CommunicationTB,
  title={Communication Topologies Between Learning Agents in Deep Reinforcement Learning},
  author={D. Adjodah and Dan Calacci and Abhimanyu Dubey and Anirudh Goyal and P. Krafft and E. M. Egido and A. Pentland},
  journal={ArXiv},
  year={2019},
  volume={abs/1902.06740}
}
  • D. Adjodah, Dan Calacci, +4 authors A. Pentland
  • Published 2019
  • Mathematics, Computer Science
  • ArXiv
  • A common technique to improve speed and robustness of learning in deep reinforcement learning (DRL) and many other machine learning algorithms is to run multiple learning agents in parallel. A neglected component in the development of these algorithms has been how best to arrange the learning agents involved to better facilitate distributed search. Here we draw upon results from the networked optimization and collective intelligence literatures suggesting that arranging learning agents in less… CONTINUE READING
    2 Citations
    Communication-Efficient Edge AI: Algorithms and Systems
    • 16
    • PDF

    References

    SHOWING 1-10 OF 30 REFERENCES
    IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
    • 524
    • PDF
    Fully Decentralized Multi-Agent Reinforcement Learning with Networked Agents
    • 160
    • PDF
    Massively Parallel Methods for Deep Reinforcement Learning
    • 286
    • PDF
    Asynchronous Methods for Deep Reinforcement Learning
    • 3,673
    • PDF
    Population Based Training of Neural Networks
    • 247
    • Highly Influential
    • PDF
    Evolution Strategies as a Scalable Alternative to Reinforcement Learning
    • 720
    • Highly Influential
    • PDF
    An Introduction to Collective Intelligence
    • 234
    • PDF