Score Normalization for Demographic Fairness in Face Recognition

• URL: http://arxiv.org/abs/2407.14087v2
• Date: Mon, 22 Jul 2024 13:59:10 GMT
• Title: Score Normalization for Demographic Fairness in Face Recognition
• Authors: Yu Linghu, Tiago de Freitas Pereira, Christophe Ecabert, Sébastien Marcel, Manuel Günther,
• Abstract summary: Well-known sample-centered score normalization techniques, Z-norm and T-norm, do not improve fairness for high-security operating points. We extend the standard Z/T-norm to integrate demographic information in normalization. We show that our techniques generally improve the overall fairness of five state-of-the-art pre-trained face recognition networks.
• Score: 16.421833444307232
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Fair biometric algorithms have similar verification performance across different demographic groups given a single decision threshold. Unfortunately, for state-of-the-art face recognition networks, score distributions differ between demographics. Contrary to work that tries to align those distributions by extra training or fine-tuning, we solely focus on score post-processing methods. As proved, well-known sample-centered score normalization techniques, Z-norm and T-norm, do not improve fairness for high-security operating points. Thus, we extend the standard Z/T-norm to integrate demographic information in normalization. Additionally, we investigate several possibilities to incorporate cohort similarities for both genuine and impostor pairs per demographic to improve fairness across different operating points. We run experiments on two datasets with different demographics (gender and ethnicity) and show that our techniques generally improve the overall fairness of five state-of-the-art pre-trained face recognition networks, without downgrading verification performance. We also indicate that an equal contribution of False Match Rate (FMR) and False Non-Match Rate (FNMR) in fairness evaluation is required for the highest gains. Code and protocols are available.

Related papers

• Mitigating Matching Biases Through Score Calibration [1.5530839016602822]
  Biased outcomes in record matching can result in unequal error rates across demographic groups, raising ethical and legal concerns. In this paper, we adapt fairness metrics traditionally applied in regression models to evaluate cumulative bias across all thresholds in record matching. We propose a novel post-processing calibration method, leveraging optimal transport theory and Wasserstein barycenters, to balance matching scores across demographic groups.
  arXiv  Detail & Related papers  (2024-11-03T21:01:40Z)
• Individual Fairness Through Reweighting and Tuning [0.23395944472515745]
  Inherent bias within society can be amplified and perpetuated by artificial intelligence (AI) systems. Recently, Graph Laplacian Regularizer (GLR) has been used as a substitute for the common Lipschitz condition to enhance individual fairness. In this work, we investigated whether defining a GLR independently on the train and target data could maintain similar accuracy.
  arXiv  Detail & Related papers  (2024-05-02T20:15:25Z)
• Distributionally Generative Augmentation for Fair Facial Attribute Classification [69.97710556164698]
  Facial Attribute Classification (FAC) holds substantial promise in widespread applications. FAC models trained by traditional methodologies can be unfair by exhibiting accuracy inconsistencies across varied data subpopulations. This work proposes a novel, generation-based two-stage framework to train a fair FAC model on biased data without additional annotation.
  arXiv  Detail & Related papers  (2024-03-11T10:50:53Z)
• Evaluating the Fairness of Discriminative Foundation Models in Computer Vision [51.176061115977774]
  We propose a novel taxonomy for bias evaluation of discriminative foundation models, such as Contrastive Language-Pretraining (CLIP) We then systematically evaluate existing methods for mitigating bias in these models with respect to our taxonomy. Specifically, we evaluate OpenAI's CLIP and OpenCLIP models for key applications, such as zero-shot classification, image retrieval and image captioning.
  arXiv  Detail & Related papers  (2023-10-18T10:32:39Z)
• Fairness-enhancing mixed effects deep learning improves fairness on in- and out-of-distribution clustered (non-iid) data [6.596656267996196]
  We introduce the Fair Mixed Effects Deep Learning (Fair MEDL) framework. Fair MEDL quantifies cluster-invariant fixed effects (FE) and cluster-specific random effects (RE) We incorporate adversarial debiasing to promote fairness across three key metrics: Equalized Odds, Demographic Parity, and Counterfactual Fairness.
  arXiv  Detail & Related papers  (2023-10-04T20:18:45Z)
• How Robust is Your Fairness? Evaluating and Sustaining Fairness under Unseen Distribution Shifts [107.72786199113183]
  We propose a novel fairness learning method termed CUrvature MAtching (CUMA) CUMA achieves robust fairness generalizable to unseen domains with unknown distributional shifts. We evaluate our method on three popular fairness datasets.
  arXiv  Detail & Related papers  (2022-07-04T02:37:50Z)
• Fair Federated Learning for Heterogeneous Face Data [10.707311210901548]
  We consider the problem of achieving fair classification in Federated Learning (FL) under data heterogeneity. Most of the approaches proposed for fair classification require diverse data that represent the different demographic groups involved. In contrast, it is common for each client to own data that represents only a single demographic group.
  arXiv  Detail & Related papers  (2021-09-06T10:44:16Z)
• Consistent Instance False Positive Improves Fairness in Face Recognition [46.55971583252501]
  Existing methods heavily rely on accurate demographic annotations. These methods are typically designed for a specific demographic group and are not general enough. We propose a false positive rate penalty loss, which mitigates face recognition bias by increasing the consistency of instance False Positive Rate.
  arXiv  Detail & Related papers  (2021-06-10T06:20:37Z)
• Balancing Biases and Preserving Privacy on Balanced Faces in the Wild [50.915684171879036]
  There are demographic biases present in current facial recognition (FR) models. We introduce our Balanced Faces in the Wild dataset to measure these biases across different ethnic and gender subgroups. We find that relying on a single score threshold to differentiate between genuine and imposters sample pairs leads to suboptimal results. We propose a novel domain adaptation learning scheme that uses facial features extracted from state-of-the-art neural networks.
  arXiv  Detail & Related papers  (2021-03-16T15:05:49Z)
• Mitigating Face Recognition Bias via Group Adaptive Classifier [53.15616844833305]
  This work aims to learn a fair face representation, where faces of every group could be more equally represented. Our work is able to mitigate face recognition bias across demographic groups while maintaining the competitive accuracy.
  arXiv  Detail & Related papers  (2020-06-13T06:43:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.