Noise Sensitivity

Noise Sensitivity is a model-level property in machine learning that characterizes the degree to which a learning algorithm's output varies in response to small, irrelevant fluctuations or random perturbations in the training data, distinguishing it from related robustness measures by specifically quantifying susceptibility to stochastic input disturbances rather than structured or adversarial changes.

779

Publications

49.3K

Citations

2.8K

Authors

983

Institutions

Publications per year

2017–2026

197

Authors

2.8K

  1. MH
    Marja Heinonen-Guzejev

    University of Helsinki, Finland

    12 publications · 598 citations

    12

    h-index

  2. DS
    Daniel Shepherd

    Auckland University of Technology, New Zealand

    13 publications · 711 citations

    12

    h-index

  3. KH
    Kauko Heikkilä

    University of Helsinki, Finland

    11 publications · 530 citations

    11

    h-index

  4. SS
    Stephen Stansfeld

    Queen Mary University of London, United Kingdom

    10 publications · 1.1K citations

    10

    h-index

  5. BG
    Barbara Griefahn

    TU Dortmund University, Germany

    8 publications · 541 citations

    8

    h-index

1–5 of 2.8K

Venues

  1. The Journal of the Acoustical Society of America

    JASA · 19 publications · 2.5K citations

    18

    h-index

  2. Noise and Health

    23 publications · 1.1K citations

    18

    h-index

  3. 15

    h-index

  4. 13

    h-index

  5. Physical Review Letters

    PRL · 13 publications · 2.2K citations

    13

    h-index

1–5