Corpus ID: 173990609

Neural Replicator Dynamics

@article{Omidshafiei2019NeuralRD,
  title={Neural Replicator Dynamics},
  author={Shayegan Omidshafiei and D. Hennes and Dustin Morrill and R. Munos and Julien P{\'e}rolat and Marc Lanctot and A. Gruslys and Jean-Baptiste Lespiau and K. Tuyls},
  journal={ArXiv},
  year={2019},
  volume={abs/1906.00190}
}
  • Shayegan Omidshafiei, D. Hennes, +6 authors K. Tuyls
  • Published 2019
  • Computer Science, Mathematics
  • ArXiv
  • Policy gradient and actor-critic algorithms form the basis of many commonly used training techniques in deep reinforcement learning. Using these algorithms in multiagent environments poses problems such as nonstationarity and instability. In this paper, we first demonstrate that standard softmax-based policy gradient can be prone to poor performance in the presence of even the most benign nonstationarity. By contrast, it is known that the replicator dynamics, a well-studied model from… CONTINUE READING
    8 Citations
    Multi-Agent Reinforcement Learning: A Selective Overview of Theories and Algorithms
    • 73
    • PDF
    DREAM: Deep Regret minimization with Advantage baselines and Model-free learning
    • 2
    • PDF
    ACCELERATED CENTRAL INFERENCE
    A survey and critique of multiagent deep reinforcement learning
    • 90
    Fictitious Play for Mean Field Games: Continuous Time Analysis and Applications
    • 2
    • PDF
    SEED RL: Scalable and Efficient Deep-RL with Accelerated Central Inference
    • 18
    • PDF
    Bounds for Approximate Regret-Matching Algorithms
    • 1
    • PDF

    References

    SHOWING 1-10 OF 96 REFERENCES
    Actor-Critic Policy Optimization in Partially Observable Multiagent Environments
    • 67
    • PDF
    Evolutionary Dynamics of Q-Learning over the Sequence Form
    • 6
    A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning
    • 221
    • PDF
    Evolutionary Dynamics of Regret Minimization
    • 14
    • PDF
    Multiagent learning using a variable learning rate
    • 722
    • Highly Influential
    • PDF
    Frequency adjusted multi-agent Q-learning
    • 61
    • PDF
    A selection-mutation model for q-learning in multi-agent systems
    • 99
    • PDF
    Addressing Environment Non-Stationarity by Repeating Q-learning Updates
    • 23
    • PDF
    Learning Through Reinforcement and Replicator Dynamics
    • 613
    • PDF
    Negative Update Intervals in Deep Multi-Agent Reinforcement Learning
    • 10
    • PDF