Deep Ensembles

Deep Ensembles is a method in uncertainty estimation that combines predictions from multiple independently trained deep neural networks to improve predictive accuracy and quantify uncertainty through model diversity.

574

Publications

38.9K

Citations

2.1K

Authors

804

Institutions

Publications per year

2017–2026

529

Authors

2.1K

  1. YG
    Yarin Gal

    University of Oxford, United Kingdom

    12 publications · 764 citations

    12

    h-index

  2. BL
    Balaji Lakshminarayanan

    Google (United States), United States

    9 publications · 4.4K citations

    9

    h-index

  3. JV
    Joost van Amersfoort

    University of Oxford, United Kingdom

    8 publications · 494 citations

    8

    h-index

  4. AG
    Andrew Gordon Wilson

    Cornell University, United States

    6 publications · 732 citations

    6

    h-index

  5. LS
    Lewis Smith

    University of Oxford, United Kingdom

    5 publications · 541 citations

    5

    h-index

1–5 of 2.1K

Venues

  1. arXiv (Cornell University)

    Repository · 103 publications · 10K citations

    41

    h-index

  2. IEEE Access

    12 publications · 583 citations

    12

    h-index

  3. Scientific Reports

    10 publications · 393 citations

    10

    h-index

  4. Pattern Recognition

    8 publications · 687 citations

    8

    h-index

  5. Multimedia Tools and Applications

    8 publications · 416 citations

    8

    h-index

1–5