# FAIR-Ensemble: When Fairness Naturally Emerges From Deep Ensembling

## TL;DR

This paper demonstrates that simple deep neural network ensembles can naturally improve fairness across subgroups, especially benefiting minority groups, even without complex modifications.

## Contribution

The study reveals that homogeneous DNN ensembles inherently promote fairness and explores the stochastic factors influencing subgroup performance.

## Key findings

- Minority group performance improves with ensemble size.
- Fairness benefits persist even with many models, e.g., 20.
- Stochastic sources like initialization and data augmentation affect fairness outcomes.

## Abstract

Ensembling multiple Deep Neural Networks (DNNs) is a simple and effective way to improve top-line metrics and to outperform a larger single model. In this work, we go beyond top-line metrics and instead explore the impact of ensembling on subgroup performances. Surprisingly, we observe that even with a simple homogeneous ensemble -- all the individual DNNs share the same training set, architecture, and design choices -- the minority group performance disproportionately improves with the number of models compared to the majority group, i.e. fairness naturally emerges from ensembling. Even more surprising, we find that this gain keeps occurring even when a large number of models is considered, e.g. $20$, despite the fact that the average performance of the ensemble plateaus with fewer models. Our work establishes that simple DNN ensembles can be a powerful tool for alleviating disparate impact from DNN classifiers, thus curbing algorithmic harm. We also explore why this is the case. We find that even in homogeneous ensembles, varying the sources of stochasticity through parameter initialization, mini-batch sampling, and data-augmentation realizations, results in different fairness outcomes.