Least-squares Temporal-difference Learning

Least-Squares Temporal-Difference Learning is a value function approximation method that minimizes the Bellman error of temporal-difference updates in a least-squares sense through linear regression, distinguishing itself from other temporal-difference algorithms by solving for parameter updates analytically rather than incrementally.

44

Publications

3.6K

Citations

90

Authors

53

Institutions

Publications per year

2017–2026

9

Authors

90

  1. MG
    Mohammad Ghavamzadeh

    Adobe Systems (United States), United States

    5 publications · 333 citations

    5

    h-index

  2. IC
    Ioannis Ch. Paschalidis

    Boston University, United States

    3 publications · 63 citations

    3

    h-index

  3. CS
    Csaba Szepesvári

    University of Alberta, Canada

    3 publications · 178 citations

    3

    h-index

  4. XX
    Xin Xu

    National University of Defense Technology, China

    3 publications · 419 citations

    3

    h-index

  5. SJ
    Steven J. Bradtke

    University of Massachusetts Amherst, United States

    2 publications · 664 citations

    2

    h-index

1–5 of 90

Venues

  1. Machine Learning

    4 publications · 1.5K citations

    4

    h-index

  2. arXiv (Cornell University)

    Repository · 3 publications · 97 citations

    3

    h-index

  3. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 2 publications · 21 citations

    2

    h-index

  4. 2

    h-index

  5. PolyPublie (École Polytechnique de Montréal)

    Repository · 2 publications · 167 citations

    2

    h-index

1–5