Noise-tolerant fair classification

By Alexandre Louis Lamy, Ziyuan Zhong, Aditya Krishna Menon, Nakul Verma. 2019

83 citations recorded by the source corpus.

Fairness-aware learning involves designing algorithms that do not discriminate with respect to some sensitive feature (e.g., race or gender). Existing work on the problem operates under the assumption that the sensitive feature available in one's training sample is perfectly reliable. This assumption may be violated in many real-world cases: for example, respondents to a survey may choose to conceal or obfuscate their group identity out of fear of potential discrimination. This poses the question of whether one can still learn fair classifiers given noisy sensitive features. In this paper, we answer the question in the affirmative: we show that if one measures fairness using the mean-difference score, and sensitive features are subject to noise from the mutually contaminated learning model, then owing to a simple identity we only need to change the desired fairness-tolerance. The requisite tolerance can be estimated by leveraging existing noise-rate estimators from the label noise literature. We finally show that our procedure is empirically effective on two case-studies involving sensitive feature censoring.

Subfields: Artificial Intelligence

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Beyond Individual and Group Fairness (2020, background): “A recent line of work has studied this estimation problem when one has limited and/or noisy access to sensitive attribute information [Gupta et al., 2018, Coston et al., 2019, Lamy et al., 2019, Wang et al., 2020].”
Data collection and quality challenges in deep learning: a data-centric AI perspective (2021, background): “In addition, noise-tolerant fair training techniques [92] have been proposed where the idea is to change the unfairness tolerance to estimate the fairness of the true data distribution.”
Robust Optimization for Fairness with Noisy Protected Groups (2020, background): “[42] showed that when there are only two protected groups, one need only tighten the “unfairness tolerance” when enforcing fairness with respect to the noisy groups.”
It's COMPASlicated: The Messy Relationship between RAI Datasets and Algorithmic Fairness Benchmarks (2021, background): “Lamy et al. [44] show that under a speciﬁc noise model, original fairness constraints can be guaranteed for the mean-difference score criterion by increasing a constraint violation parameter.”