Concepedia

Optimistic Policy Iteration

Optimistic Policy Iteration is a variant of policy iteration in reinforcement learning where the policy evaluation step is truncated after a single iteration, allowing the algorithm to proceed to policy improvement based on an incomplete but optimistic estimate of the value function, thereby accelerating convergence in certain settings.

20

Publications

576

Citations

42

Authors

29

Institutions

Publications per year

2017–2026

10

Variants

Modified Policy Iteration

Parents

Publications

20

Provably Efficient Exploration in Policy Optimization

Qi Cai, Zhuoran Yang, Chi Jin et al. · 2020 · 86 citations

, , +15

Error Bounds of Adaptive Dynamic Programming Algorithms for Solving Undiscounted Optimal Control Problems

Derong Liu, Hongliang Li, Ding Wang · IEEE Transactions on Neural Networks and Learning Systems · 2015 · 70 citations

, , +17

Reinforcement Learning for Adaptive Optimal Stationary Control of Linear Stochastic Systems

Bo Pang, Zhong‐Ping Jiang · IEEE Transactions on Automatic Control · 2022 · 48 citations

, , +17

Model-free reinforcement learning as mixture learning

Nikos Vlassis, Marc Toussaint · 2009 · 42 citations · Full text

, , +16

Convergence of Optimistic and Incremental Q-Learning

Eyal Even-Dar, Yishay Mansour · 2001 · 41 citations

, , +15

1–5 of 20

Authors

42

  1. ZY
    Zhuoran Yang

    Princeton University, United States

    4 publications · 144 citations

    4

    h-index

  2. CJ
    Chi Jin

    University of California, Berkeley, United States

    3 publications · 118 citations

    3

    h-index

  3. DP
    Daniel Polani

    University of Hertfordshire, United Kingdom

    2 publications · 49 citations

    2

    h-index

  4. ZJ
    Zhong‐Ping Jiang

    New York University, United States

    2 publications · 84 citations

    2

    h-index

  5. ZW
    Zhaoran Wang

    Princeton University, United States

    2 publications · 103 citations

    2

    h-index

1–5 of 42

Institutions

29

  1. Princeton University

    Princeton, United States

    2 h-index

    4 publications

    206 citations

  2. Johannes Gutenberg University Mainz

    Mainz, Germany

    2 h-index

    2 publications

    49 citations

  3. New York University

    New York, United States

    2 h-index

    4 publications

    168 citations

  4. Tel Aviv University

    Tel Aviv, Israel

    1 h-index

    2 publications

    82 citations

  5. Shandong Institute of Automation

    Jinan, China

    1 h-index

    3 publications

    210 citations

  6. Centre National de la Recherche Scientifique

    Paris, France

    1 h-index

    1 publications

    18 citations

  7. University of Wisconsin–Madison

    Madison, United States

    1 h-index

    2 publications

    52 citations

  8. University of Oxford

    Oxford, United Kingdom

    1 h-index

    1 publications

    10 citations

1–8 of 29

Venues

  1. arXiv (Cornell University)

    Repository · 4 publications · 68 citations

    4

    h-index

  2. 1

    h-index

  3. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 1 publications · 36 citations

    1

    h-index

  4. IEEE Transactions on Automatic Control

    1 publications · 48 citations

    1

    h-index

  5. Computers & Mathematics with Applications

    1 publications · 18 citations

    1

    h-index

1–5