Despite growing empirical evidence of bias amplification in machine learning, its theoretical foundations remain poorly understood. We develop a formal framework for majority-minority learning tasks, showing how standard training can favor majority groups and produce stereotypical predictors that neglect minority-specific features. Assuming population and variance imbalance, our analysis reveals three key findings: (i) the close proximity between ``full-data'' and stereotypical predictors, (ii) the dominance of a region where training the entire model tends to merely learn the majority traits, and (iii) a lower bound on the additional training required. Our results are illustrated through experiments in deep learning for tabular and image classification tasks.
View on arXiv@article{bachoc2025_2505.13122,
title={ When majority rules, minority loses: bias amplification of gradient descent },
author={ François Bachoc and Jérôme Bolte and Ryan Boustany and Jean-Michel Loubes },
journal={arXiv preprint arXiv:2505.13122},
year={ 2025 }
}