Concepedia

Overparameterized Models

Overparameterized Models is a machine learning model whose number of trainable parameters significantly exceeds the number of training samples, distinguishing it from underparameterized models by its capacity to perfectly fit or interpolate the training data, often leading to concerns about generalization despite high empirical performance.

182

Publications

16.2K

Citations

407

Authors

159

Institutions

Publications per year

2017–2026

142

Parents

Publications

182

On hyperparameter optimization of machine learning algorithms: Theory and practice

Li Yang, Abdallah Shami · Neurocomputing · 2020 · 3.1K citations · Full text

, , +11

Hyperparameter Optimization for Machine Learning Models Based on Bayesian Optimization

Jia Wu, Xiu Yun Chen, Hao Zhang et al. · HAL (Le Centre pour la Communication Scientifique Directe) · 2019 · 1.4K citations · Full text

, , +15

Model selection and overfitting

Jake Lever, Martin Krzywinski, Naomi Altman · Nature Methods · 2016 · 700 citations

, , +11

The Marginal Value of Adaptive Gradient Methods in Machine Learning

Ashia Wilson, Rebecca Roelofs, Mitchell Stern et al. · arXiv (Cornell University) · 2017 · 553 citations · Full text

, , +16

Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization

Shiori Sagawa, Pang Wei Koh, Tatsunori Hashimoto et al. · arXiv (Cornell University) · 2019 · 364 citations · Full text

, , +16

1–5 of 182

Authors

407

  1. MS
    Mahdi Soltanolkotabi

    University of Southern California, United States

    8 publications · 573 citations

    8

    h-index

  2. MB
    Mikhail Belkin

    The Ohio State University, United States

    7 publications · 326 citations

    7

    h-index

  3. SO
    Samet Oymak

    California Institute of Technology, United States

    7 publications · 567 citations

    7

    h-index

  4. QG
    Quanquan Gu

    University of California, Los Angeles, United States

    5 publications · 366 citations

    5

    h-index

  5. PL
    Percy Liang

    Stanford University, United States

    5 publications · 630 citations

    5

    h-index

1–5 of 407

Institutions

159

  1. University of California, Berkeley

    Berkeley, United States

    11 h-index

    19 publications

    1.6K citations

  2. Google (United States)

    Mountain View, United States

    10 h-index

    21 publications

    1.2K citations

  3. Princeton University

    Princeton, United States

    9 h-index

    12 publications

    635 citations

  4. Stanford University

    Stanford, United States

    7 h-index

    15 publications

    1.2K citations

  5. University of Southern California

    Los Angeles, United States

    6 h-index

    10 publications

    348 citations

  6. University of California, Los Angeles

    Los Angeles, United States

    5 h-index

    14 publications

    1.2K citations

  7. University of California San Diego

    San Diego, United States

    4 h-index

    4 publications

    124 citations

  8. University of Wisconsin–Madison

    Madison, United States

    4 h-index

    8 publications

    620 citations

1–8 of 159

Venues

  1. arXiv (Cornell University)

    Repository · 78 publications · 4.9K citations

    36

    h-index

  2. Neural Information Processing Systems

    Conference · 5 publications · 439 citations

    5

    h-index

  3. IEEE Transactions on Automatic Control

    3 publications · 42 citations

    3

    h-index

  4. IEEE Journal on Selected Areas in Information Theory

    JSAIT · 2 publications · 168 citations

    2

    h-index

  5. Neurocomputing

    2 publications · 3.2K citations

    2

    h-index

1–5