Metric-Free Individual Fairness in Online Learning

TL;DR

This paper introduces a method for online learning that ensures individual fairness without knowing the similarity measure, using an auditor to detect violations and achieving strong regret and fairness guarantees.

Contribution

It presents a novel reduction framework for online classification under unknown fairness constraints, enabling the use of existing algorithms to ensure fairness and accuracy.

Findings

Achieves sub-linear regret and fairness violations in online learning.

Establishes PAC-style fairness and accuracy guarantees in the stochastic setting.

Shows online learning under unknown fairness constraints is feasible without parametric similarity measures.

Abstract

We study an online learning problem subject to the constraint of individual fairness, which requires that similar individuals are treated similarly. Unlike prior work on individual fairness, we do not assume the similarity measure among individuals is known, nor do we assume that such measure takes a certain parametric form. Instead, we leverage the existence of an auditor who detects fairness violations without enunciating the quantitative measure. In each round, the auditor examines the learner's decisions and attempts to identify a pair of individuals that are treated unfairly by the learner. We provide a general reduction framework that reduces online classification in our model to standard online classification, which allows us to leverage existing online learning algorithms to achieve sub-linear regret and number of fairness violations. Surprisingly, in the stochastic setting where the data are drawn independently from a distribution, we are also able to establish PAC-style fairness and accuracy generalization guarantees (Rothblum and Yona [2018]), despite only having access to a very restricted form of fairness feedback. Our fairness generalization bound qualitatively matches the uniform convergence bound of Rothblum and Yona [2018], while also providing a meaningful accuracy generalization guarantee. Our results resolve an open question by Gillen et al. [2018] by showing that online learning under an unknown individual fairness constraint is possible even without assuming a strong parametric form of the underlying similarity measure.