Bandit Algorithms

Bandit Algorithms is a class of sequential decision-making algorithms that address the exploration-exploitation tradeoff in stochastic environments by iteratively selecting actions to maximize cumulative rewards based on partial feedback.

1.9K

Publications

112.1K

Citations

3.1K

Authors

821

Institutions

Publications per year

2017–2026

942

Variants

Mab AlgorithmsMulti-armed Bandit Algorithms

Authors

3.1K

  1. CS
    Csaba Szepesvári

    University of Alberta, Canada

    42 publications · 3.7K citations

    27

    h-index

  2. AS
    Aleksandrs Slivkins

    Microsoft (United States), United States

    35 publications · 2.3K citations

    23

    h-index

  3. SB
    Sébastien Bubeck

    Microsoft (United States), United States

    28 publications · 2.2K citations

    21

    h-index

  4. JL
    John Langford

    Microsoft (United States), United States

    26 publications · 5.5K citations

    21

    h-index

  5. RM
    Rémi Munos

    Google (United States), United States

    28 publications · 2.9K citations

    21

    h-index

1–5 of 3.1K

Venues

  1. arXiv (Cornell University)

    Repository · 435 publications · 19.2K citations

    62

    h-index

  2. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 58 publications · 2.2K citations

    26

    h-index

  3. Neural Information Processing Systems

    Conference · 40 publications · 1.4K citations

    20

    h-index

  4. IEEE Transactions on Automatic Control

    22 publications · 1.9K citations

    19

    h-index

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

    Repository · 21 publications · 1.8K citations

    18

    h-index

1–5