Successor Representation

Successor Representation is a predictive value function that belongs to the class of model-based reinforcement learning representations and is distinguished by its encoding of expected future state occupancy under a given policy, thereby bridging model-free and model-based value prediction through a learned transition structure.

52

Publications

4.3K

Citations

98

Authors

56

Institutions

Publications per year

2017–2026

36

Variants

Sr

Authors

98

  1. SJ
    Samuel J. Gershman

    Harvard University, United States

    7 publications · 1.3K citations

    7

    h-index

  2. ND
    Nathaniel D. Daw

    Princeton University, United States

    6 publications · 913 citations

    6

    h-index

  3. IM
    Ida Momennejad

    Princeton University, United States

    6 publications · 963 citations

    6

    h-index

  4. ST
    Son Thanh To

    New Mexico State University, United States

    5 publications · 97 citations

    5

    h-index

  5. EP
    Enrico Pontelli

    New Mexico State University, United States

    5 publications · 97 citations

    5

    h-index

1–5 of 98

Venues

  1. arXiv (Cornell University)

    Repository · 14 publications · 576 citations

    12

    h-index

  2. bioRxiv (Cold Spring Harbor Laboratory)

    Repository · 6 publications · 232 citations

    6

    h-index

  3. eLife

    5 publications · 634 citations

    5

    h-index

  4. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 3 publications · 40 citations

    3

    h-index

  5. 3

    h-index

1–5