Spectral Algorithms for Computing Fair Support Vector Machines

By Matt Olfat, Anil Jayanti Aswani. 2017

36 citations recorded by the source corpus.

Classifiers and rating scores are prone to implicitly codifying biases, which may be present in the training data, against protected classes (i.e., age, gender, or race). So it is important to understand how to design classifiers and scores that prevent discrimination in predictions. This paper develops computationally tractable algorithms for designing accurate but fair support vector machines (SVM's). Our approach imposes a constraint on the covariance matrices conditioned on each protected class, which leads to a nonconvex quadratic constraint in the SVM formulation. We develop iterative algorithms to compute fair linear and kernel SVM's, which solve a sequence of relaxations constructed using a spectral decomposition of the nonconvex constraint. Its effectiveness in achieving high prediction accuracy while ensuring fairness is shown through numerical experiments on several data sets.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Statistics and Probability

How researchers see this paper

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

In Computational Mechanics

In Computer Vision and Pattern Recognition

In Statistics and Probability

Citations in the corpus

Empirical Risk Minimization under Fairness Constraints (2018, background): “tate-of-the-art solutions and do not provide consistency properties except for the [1] which, contrarily to our proposal, does not enforce a fairness constraint directly in the learning phase and the [30] which proposes a computational tractable fair SVM starting from a constraint on the covariance ”
Convex Formulations for Fair Principal Component Analysis (2018, methodology): “We compare these all with the method of Calmon et al., as before, but we also compare to the Fair SVM (FSVM) method of (Olfat and Aswani 2017) (run with hyperparam-eters δ = 0 , µ = 0 .”
Explainable Deep Learning: A Field Guide for the Uninitiated (2020, background): “…al., 2017a; Woodworth et al., 2017; Zafar et al., 2017b; Bechavod & Ligett, 2017; Kearns et al., 2018; Pérez-Suay et al., 2017; Berk et al., 2017; Olfat & Aswani, 2018; Agarwal et al., 2018; Menon & Williamson, 2018; Donini et al., 2018; Dwork et al., 2018) directly introduce fairness learning…”
Fairness in Machine Learning (2020, unknown): “…21, 22, 23, 25, 28, 29, 31, 34, 44, 45, 46, 52, 53, 54, 55, 56, 59, 66, 69, 71, 72, 73, 74, 75, 82, 84, 87, 89, 90, 94, 95, 97, 98, 99, 103, 104, 106, 114, 115, 116, 121, 123, 126, 138, 139, 141, 148, 149, 153, 156, 159, 165, 174, 176, 177, 178, 179, 180, 182, 183, 184, 186, 187, 191, 192, 193].”
Fairness risk measures (2019, background): “An alternate measure that is similarly general, but convex, is the covariance between the target and sensitive features [Zafar et al., 2017c, Olfat and Aswani, 2018, Donini et al., 2018]:”