Concepedia

Policy Gradients

Policy Gradients is a reinforcement learning method that directly optimizes the policy by adjusting its parameters in the direction of the gradient of expected cumulative reward, distinguishing it from value-based methods by operating in the policy space rather than relying on value function approximations.

589

Publications

52.9K

Citations

1.5K

Authors

465

Institutions

Publications per year

2017–2026

485

Variants

Policy Gradient Methods

Parents

Publications

589

Simple statistical gradient-following algorithms for connectionist reinforcement learning

Ronald J. Williams · Machine Learning · 1992 · 7.5K citations · Full text

, , +11

Policy Gradient Methods for Reinforcement Learning with Function Approximation

Richard S. Sutton, David McAllester, Satinder Singh et al. · 1999 · 5K citations

, , +13

High-Dimensional Continuous Control Using Generalized Advantage Estimation

John Schulman, Philipp Moritz, Sergey Levine et al. · arXiv (Cornell University) · 2015 · 1.8K citations · Full text

, , +17

Deterministic policy gradient algorithms

David Silver, Guy Lever, Nicolas Heess et al. · HAL (Le Centre pour la Communication Scientifique Directe) · 2014 · 1.7K citations · Full text

, , +16

Counterfactual Multi-Agent Policy Gradients

Jakob Foerster, Gregory Farquhar, Triantafyllos Afouras et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 2018 · 1.6K citations · Full text

, , +17

1–5 of 589

Authors

1.5K

  1. PA
    Pieter Abbeel

    University of California, Berkeley, United States

    11 publications · 2.9K citations

    10

    h-index

  2. SL
    Sergey Levine

    University of California, Berkeley, United States

    9 publications · 2.6K citations

    9

    h-index

  3. ZY
    Zhuoran Yang

    Princeton University, United States

    9 publications · 267 citations

    9

    h-index

  4. RM
    Rémi Munos

    Google (United States), United States

    9 publications · 348 citations

    9

    h-index

  5. JP
    Jan Peters

    Technische Universität Darmstadt, Germany

    8 publications · 2.4K citations

    9

    h-index

1–5 of 1.5K

Institutions

465

  1. Google (United States)

    Mountain View, United States

    18 h-index

    72 publications

    6.3K citations

  2. University of Oxford

    Oxford, United Kingdom

    13 h-index

    32 publications

    11.4K citations

  3. Princeton University

    Princeton, United States

    12 h-index

    21 publications

    714 citations

  4. University of Alberta

    Edmonton, Canada

    11 h-index

    23 publications

    1.5K citations

  5. South China University of Technology

    Guangzhou, China

    10 h-index

    19 publications

    817 citations

  6. Massachusetts Institute of Technology

    Cambridge, United States

    10 h-index

    33 publications

    2.1K citations

  7. University of California, Berkeley

    Berkeley, United States

    10 h-index

    23 publications

    9.5K citations

  8. Tsinghua University

    Beijing, China

    10 h-index

    22 publications

    943 citations

1–8 of 465

Venues

  1. arXiv (Cornell University)

    Repository · 154 publications · 11.9K citations

    48

    h-index

  2. IEEE Access

    18 publications · 743 citations

    17

    h-index

  3. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 13 publications · 2.5K citations

    13

    h-index

  4. 10

    h-index

  5. Sensors

    7 publications · 239 citations

    7

    h-index

1–5