Bachoc, François, Bolte, Jérôme, Boustany, Ryan and Loubès, Jean-Michel (2025) When majority rules, minority loses: bias amplification of gradient descent. TSE Working Paper, n. 25-1641, Toulouse

Preview

Text Download (1MB) | Preview

Abstract

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.