Prioritized Sweeping

Prioritized Sweeping is a model-based reinforcement learning algorithm that belongs to the class of dynamic programming methods, distinguished by its use of a priority queue to selectively update state values based on the magnitude of their backed-up errors, thereby accelerating convergence in sparse environments.

17

Publications

1.6K

Citations

42

Authors

21

Institutions

Publications per year

2017–2026

2

Authors

42

  1. RS
    Richard S. Sutton

    University of Alberta, Canada

    3 publications · 145 citations

    3

    h-index

  2. CG
    Christopher G. Atkeson

    Carnegie Mellon University, United States

    3 publications · 1.1K citations

    3

    h-index

  3. AM
    Andrew Moore

    Carnegie Mellon University, United States

    3 publications · 1.1K citations

    3

    h-index

  4. HV
    Harm van Seijen

    University of Alberta, Canada

    2 publications · 38 citations

    2

    h-index

  5. MG
    Mevludin Glavić

    University of Liège, Belgium

    2 publications · 40 citations

    2

    h-index

1–5 of 42

Venues

  1. arXiv (Cornell University)

    Repository · 2 publications · 132 citations

    2

    h-index

  2. Machine Learning

    2 publications · 1.1K citations

    2

    h-index

  3. 1

    h-index

  4. IEEE Transactions on Vehicular Technology

    1 publications · 62 citations

    1

    h-index

  5. IEEE/ASME Transactions on Mechatronics

    1 publications · 70 citations

    1

    h-index

1–5