Algorithmic Discrimination: Formulation and Exploration in Deep Learning-based Face Biometrics

TL;DR

This paper investigates demographic biases in deep learning-based face recognition, revealing significant performance disparities across groups and proposing a general framework for understanding algorithmic discrimination.

Contribution

It introduces a comprehensive discrimination-aware framework for face biometrics and demonstrates demographic biases in popular deep face recognition models using the DiveFace database.

Findings

Demographic groups in popular datasets cause biases in face recognition performance.

Pre-trained deep models show strong discrimination at the feature space level.

Performance varies significantly across demographic groups.

Abstract

The most popular face recognition benchmarks assume a distribution of subjects without much attention to their demographic attributes. In this work, we perform a comprehensive discrimination-aware experimentation of deep learning-based face recognition. The main aim of this study is focused on a better understanding of the feature space generated by deep models, and the performance achieved over different demographic groups. We also propose a general formulation of algorithmic discrimination with application to face biometrics. The experiments are conducted over the new DiveFace database composed of 24K identities from six different demographic groups. Two popular face recognition models are considered in the experimental framework: ResNet-50 and VGG-Face. We experimentally show that demographic groups highly represented in popular face databases have led to popular pre-trained deep face models presenting strong algorithmic discrimination. That discrimination can be observed both qualitatively at the feature space of the deep models and quantitatively in large performance differences when applying those models in different demographic groups, e.g. for face biometrics.