When do Minimax-fair Learning and Empirical Risk Minimization Coincide?

Harvineet Singh, Matthäus Kleindessner, Volkan Cevher, Rumi Chunara, Chris Russell

Research output: Contribution to journalConference articlepeer-review


Minimax-fair machine learning minimizes the error for the worst-off group. However, empirical evidence suggests that when sophisticated models are trained with standard empirical risk minimization (ERM), they often have the same performance on the worst-off group as a minimax-trained model. Our work makes this counter-intuitive observation concrete. We prove that if the hypothesis class is sufficiently expressive and the group information is recoverable from the features, ERM and minimax-fairness learning formulations indeed have the same performance on the worst-off group. We provide additional empirical evidence of how this observation holds on a wide range of datasets and hypothesis classes. Since ERM is fundamentally easier than minimax optimization, our findings have implications on the practice of fair machine learning.

Original languageEnglish (US)
Pages (from-to)31969-31989
Number of pages21
JournalProceedings of Machine Learning Research
StatePublished - 2023
Event40th International Conference on Machine Learning, ICML 2023 - Honolulu, United States
Duration: Jul 23 2023Jul 29 2023

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability


Dive into the research topics of 'When do Minimax-fair Learning and Empirical Risk Minimization Coincide?'. Together they form a unique fingerprint.

Cite this