Privacy-Preserving Race/Ethnicity Estimation for Algorithmic Bias Measurement in the U.S
This work addresses the problem of algorithmic bias measurement for AI practitioners and platforms, but it is incremental as it builds on prior methods like BISG and adds privacy protections.
The paper tackles the challenge of enabling AI fairness measurements for race/ethnicity among U.S. LinkedIn members by developing a privacy-preserving method that combines existing models with privacy-enhancing technologies, allowing for meaningful bias assessments without compromising member privacy.
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.