Systems incorporating biometric technologies have become ubiquitous in personal, commercial, and governmental identity management applications. Both cooperative (e.g., access control) and noncooperative (e.g., surveillance and forensics) systems have benefited from biometrics. Such systems rely on the uniqueness of certain biological or behavioral characteristics of human beings, which enable for individuals to be reliably recognized using automated algorithms. Recently, however, there has been a wave of public and academic concerns regarding the existence of systemic bias in automated decision systems (including biometrics). Most prominently, face recognition algorithms have often been labeled as “racist” or “biased” by the media, nongovernmental organizations, and researchers alike. The main contributions of this article are: 1) an overview of the topic of algorithmic bias in the context of biometrics; 2) a comprehensive survey of the existing literature on biometric bias estimation and mitigation; 3) a discussion of the pertinent technical and social matters; and 4) an outline of the remaining challenges and future work items, both from technological and social points of view.

Demographic Bias in Biometrics: A Survey on an Emerging Challenge

Systems incorporating biometric technologies have become ubiquitous in personal, commercial, and governmental identity management applications. Both cooperative (e.g. access control) and non-cooperative (e.g. surveillance and forensics) systems have benefited from biometrics. Such systems rely on the uniqueness of certain biological or behavioural characteristics of human beings, which enable for individuals to be reliably recognised using automated algorithms. 
Recently, however, there has been a wave of public and academic concerns regarding the existence of systemic bias in automated decision systems (including biometrics). Most prominently, face recognition algorithms have often been labelled as "racist" or "biased" by the media, non-governmental organisations, and researchers alike. 
The main contributions of this article are: (1) an overview of the topic of algorithmic bias in the context of biometrics, (2) a comprehensive survey of the existing literature on biometric bias estimation and mitigation, (3) a discussion of the pertinent technical and social matters, and (4) an outline of the remaining challenges and future work items, both from technological and social points of view.

Demographic Bias in Biometrics: A Survey on an Emerging Challenge.

Beating the bias in facial recognition technology

Recent studies show that there exist significant racial bias among state-of-the-art (SOTA) face recognition algorithms, i.e., the accuracy for Caucasian is consistently higher than that for other races like African and Asian. To mitigate racial bias, we propose the race adaptive margin based face recognition (RamFace) model, designed under the multi-task learning framework with the race classification as the auxiliary task. The experiments show that the race classification task can enforce the model to learn the racial features and thus improve the discriminability of the extracted feature representations. In addition, a racial bias robust loss function, i.e., race adaptive margin loss, is proposed such that different optimal margins can be automatically derived for different races in training the model, which further mitigates the racial bias. The experimental results show that on RFW dataset, our model not only achieves SOTA face recognition accuracy but also mitigates the racial bias problem. Besides, RamFace is also tested on several public face recognition evaluation benchmarks, i.e., LFW, CPLFW and CALFW, and achieves better performance than the commonly used face recognition methods, which justifies the generalization capability of RamFace.

RamFace: Race Adaptive Margin Based Face Recognition for Racial Bias Mitigation

Face presentation attack detection (PAD) is critical to secure face recognition (FR) applications from presentation attacks. FR performance has been shown to be unfair to certain demographic and non-demographic groups. However, the fairness of face PAD is an understudied issue, mainly due to the lack of appropriately annotated data. To address this issue, this work ﬁrst presents a Combined Attribute Annotated PAD Dataset (CAAD-PAD) by combining several well-known PAD datasets where we provide seven human-annotated attribute labels. This work then comprehensively analyses the fairness of a set of face PADs and its relation to the nature of training data and the Operational Decision Threshold Assignment (ODTA) on different data groups by studying four face PAD approaches on our CAAD- PAD. To simultaneously represent both the PAD fairness and the absolute PAD performance, we introduce a novel metric, namely the Accuracy Balanced Fairness (ABF). Extensive experiments on CAAD-PAD show that the training data and ODTA induce unfairness on gender, occlusion, and other attribute groups. Based on these analyses, we propose a data augmentation method, FairSWAP, which aims to disrupt the identity/semantic information and guide models to mine attack cues rather than attribute- related information. Detailed experimental results demonstrate that FairSWAP generally enhances both the PAD performance and the fairness of face PAD.

Fairness in Face Presentation Attack Detection

As face recognition algorithms become more accurate and get deployed more widely, it becomes increasingly important to ensure that the algorithms work equally well for everyone. We study the geographic performance differentials—differences in false acceptance and false rejection rates across different countries—when comparing selfies against photos from ID documents. We show how to mitigate geographic performance differentials using sampling strategies despite large imbalances in the dataset. Using vanilla domain adaptation strategies to fine-tune a face recognition CNN on domain-specific doc-selfie data improves the performance of the model on such data, but, in the presence of imbalanced training data, also significantly increases the demographic bias. We then show how to mitigate this effect by employing sampling strategies to balance the training procedure.

/pdf/reducing-geographic-performance-differentials-for-face-1n20tnf8rj.pdf

Reducing Geographic Performance Differentials for Face Recognition

As face recognition algorithms become more accurate and get deployed more widely, it becomes increasingly important to ensure that the algorithms work equally well for everyone. We study the geographic performance differentials-differences in false acceptance and false rejection rates across different countries-when comparing selfies against photos from ID documents. We show how to mitigate geographic performance differentials using sampling strategies despite large imbalances in the dataset. Using vanilla domain adaptation strategies to fine-tune a face recognition CNN on domain-specific doc-selfie data improves the performance of the model on such data, but, in the presence of imbalanced training data, also significantly increases the demographic bias. We then show how to mitigate this effect by employing sampling strategies to balance the training procedure.

Reducing Geographic Performance Differential for Face Recognition

The present invention relates to a machine learning inference system and processing modules thereof. In particular, the present invention relates to a machine learning inference system, a confidence module, a data minder module, a data remapping module, an adversarial defence module, and an update module. The machine learning inference system and processing modules thereof are useful for mission-critical applications to increase and maintain performance of a machine learning model.

Mohan Mahadevan

Papers

Reducing Geographic Performance Differentials for Face Recognition

Reducing Geographic Performance Differential for Face Recognition

Machine learning inference system