Privacy-Preserving Race/Ethnicity Estimation for Algorithmic Bias Measurement in the U.S

TL;DR

This paper introduces PPRE, a privacy-preserving method combining Bayesian modeling and privacy tech to estimate race/ethnicity for fairness assessments on LinkedIn, addressing privacy and legal challenges.

Contribution

The paper presents a novel privacy-preserving approach for race/ethnicity estimation using Bayesian models, survey data, and privacy technologies for fairness measurement.

Findings

PPRE enables meaningful fairness measurement while preserving privacy.

The method combines Bayesian models with secure computation and differential privacy.

Sample operations demonstrate practical applicability of the approach.

Abstract

AI fairness measurements, including tests for equal treatment, often take the form of disaggregated evaluations of AI systems. Such measurements are an important part of Responsible AI operations. These measurements compare system performance across demographic groups or sub-populations and typically require member-level demographic signals such as gender, race, ethnicity, and location. However, sensitive member-level demographic attributes like race and ethnicity can be challenging to obtain and use due to platform choices, legal constraints, and cultural norms. In this paper, we focus on the task of enabling AI fairness measurements on race/ethnicity for \emph{U.S. LinkedIn members} in a privacy-preserving manner. We present the Privacy-Preserving Probabilistic Race/Ethnicity Estimation (PPRE) method for performing this task. PPRE combines the Bayesian Improved Surname Geocoding (BISG) model, a sparse LinkedIn survey sample of self-reported demographics, and privacy-enhancing technologies like secure two-party computation and differential privacy to enable meaningful fairness measurements while preserving member privacy. We provide details of the PPRE method and its privacy guarantees. We then illustrate sample measurement operations. We conclude with a review of open research and engineering challenges for expanding our privacy-preserving fairness measurement capabilities.