Average Individual Fairness: Algorithms, Generalization and Experiments

Abstract

We propose a new family of fairness definitions for classification problems that combine some of the best properties of both statistical and individual notions of fairness. We posit not only a distribution over individuals, but also a distribution over (or collection of) classification tasks. We then ask that standard statistics (such as error or false positive/negative rates) be (approximately) equalized across individuals, where the rate is defined as an expectation over the classification tasks. Because we are no longer averaging over coarse groups (such as race or gender), this is a semantically meaningful individual-level constraint. Given a sample of individuals and classification problems, we design an oracle-efficient algorithm (i.e. one that is given access to any standard, fairness-free learning heuristic) for the fair empirical risk minimization task. We also show that given sufficiently many samples, the ERM solution generalizes in two directions: both to new individuals, and to new classification tasks, drawn from their corresponding distributions. Finally we implement our algorithm and empirically verify its effectiveness.

References in corpus (1)

• Equality of Opportunity in Supervised Learning

Cited by in corpus (5)

• A survey on datasets for fairness-aware machine learning
• Toward Operationalizing Pipeline-aware ML Fairness: A Research Agenda for Developing Practical Guidelines and Tools
• Fairness in Forecasting of Observations of Linear Dynamical Systems
• Gerrymandering Individual Fairness
• On the Fairness of Randomized Trials for Recommendation with Heterogeneous Demographics and Beyond