English

How does overparametrization affect performance on minority groups?

Machine Learning 2022-06-09 v1 Statistics Theory Statistics Theory

Abstract

The benefits of overparameterization for the overall performance of modern machine learning (ML) models are well known. However, the effect of overparameterization at a more granular level of data subgroups is less understood. Recent empirical studies demonstrate encouraging results: (i) when groups are not known, overparameterized models trained with empirical risk minimization (ERM) perform better on minority groups; (ii) when groups are known, ERM on data subsampled to equalize group sizes yields state-of-the-art worst-group-accuracy in the overparameterized regime. In this paper, we complement these empirical studies with a theoretical investigation of the risk of overparameterized random feature models on minority groups. In a setting in which the regression functions for the majority and minority groups are different, we show that overparameterization always improves minority group performance.

Keywords

Cite

@article{arxiv.2206.03515,
  title  = {How does overparametrization affect performance on minority groups?},
  author = {Subha Maity and Saptarshi Roy and Songkai Xue and Mikhail Yurochkin and Yuekai Sun},
  journal= {arXiv preprint arXiv:2206.03515},
  year   = {2022}
}
R2 v1 2026-06-24T11:42:37.975Z