Corpus ID: 219792060

Competitive Policy Optimization

@article{Prajapat2020CompetitivePO,
  title={Competitive Policy Optimization},
  author={Manish Prajapat and Kamyar Azizzadenesheli and Alexander Liniger and Yisong Yue and Anima Anandkumar},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.10611}
}
  • Manish Prajapat, Kamyar Azizzadenesheli, +2 authors Anima Anandkumar
  • Published 2020
  • Computer Science, Mathematics
  • ArXiv
  • A core challenge in policy optimization in competitive Markov decision processes is the design of efficient optimization methods with desirable convergence and stability properties. To tackle this, we propose competitive policy optimization (CoPO), a novel policy gradient approach that exploits the game-theoretic nature of competitive games to derive policy updates. Motivated by the competitive gradient optimization method, we derive a bilinear approximation of the game objective. In contrast… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 73 REFERENCES

    Competitive gradient descent, 2019

    • Florian Schäfer, Anima Anandkumar
    • 2019
    VIEW 8 EXCERPTS
    HIGHLY INFLUENTIAL

    Competitive Markov Decision Processes

    VIEW 25 EXCERPTS
    HIGHLY INFLUENTIAL

    Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments

    VIEW 14 EXCERPTS
    HIGHLY INFLUENTIAL

    Generative Adversarial Nets

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning with Opponent-Learning Awareness

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning with opponent-learning awareness. CoRR, abs/1709.04326

    • Jakob N Foerster, Richard Y Chen, +3 authors Igor Mordatch
    • 2017
    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL