Fitted Value Iteration

Fitted Value Iteration is a model-free reinforcement learning algorithm that approximates the value function by iteratively refining a parametric or non-parametric function fit to sampled Bellman backups, thereby enabling scalable and continuous-state policy evaluation.

19

Publications

1.2K

Citations

53

Authors

34

Institutions

Publications per year

2017–2026

3

Variants

Fvi

Authors

53

  1. CS
    Csaba Szepesvári

    University of Alberta, Canada

    3 publications · 361 citations

    3

    h-index

  2. MG
    Matthieu Geist

    Google (United States), United States

    3 publications · 114 citations

    3

    h-index

  3. RM
    Rémi Munos

    Google (United States), United States

    3 publications · 361 citations

    3

    h-index

  4. MG
    Mohammad Ghavamzadeh

    Adobe Systems (United States), United States

    2 publications · 97 citations

    2

    h-index

  5. BS
    Bruno Scherrer

    Institut Élie Cartan de Lorraine, France

    2 publications · 97 citations

    2

    h-index

1–5 of 53

Venues

  1. arXiv (Cornell University)

    Repository · 2 publications · 48 citations

    2

    h-index

  2. Journal of Artificial Intelligence Research

    JAIR · 1 publications · 34 citations

    1

    h-index

  3. IEEE Transactions on Smart Grid

    1 publications · 339 citations

    1

    h-index

  4. HAL (Le Centre pour la Communication Scientifique Directe)

    Repository · 1 publications · 71 citations

    1

    h-index

  5. TUbilio (Technical University of Darmstadt)

    Repository · 1 publications · 12 citations

    1

    h-index

1–5