“You Can’t Fix What You Can’t Measure”: Privately Measuring Demographic Performance Disparities in Federated Learning

Marc Juarez, Aleksandra Korolova
Proceedings of the Workshop on Algorithmic Fairness through the Lens of Causality and Privacy, PMLR 214:67-85, 2023.

Abstract

As in traditional machine learning models, models trained with federated learning may exhibit disparate performance across demographic groups. Model holders must identify these disparities to mitigate undue harm to the groups. However, measuring a model’s performance in a group requires access to information about group membership which, for privacy reasons, often has limited availability. We propose novel locally differentially private mechanisms to measure differences in performance across groups while protecting the privacy of group membership. To analyze the effectiveness of the mechanisms, we bound their error in estimating a disparity when optimized for a given privacy budget. Our results show that the error rapidly decreases for realistic numbers of participating clients, demonstrating that, contrary to what prior work suggested, protecting privacy is not necessarily in conflict with identifying performance disparities of federated models.

Cite this Paper

BibTeX
@InProceedings{pmlr-v214-juarez23a, title = {“You Can’t Fix What You Can’t Measure”: Privately Measuring Demographic Performance Disparities in Federated Learning}, author = {Juarez, Marc and Korolova, Aleksandra}, booktitle = {Proceedings of the Workshop on Algorithmic Fairness through the Lens of Causality and Privacy}, pages = {67--85}, year = {2023}, editor = {Dieng, Awa and Rateike, Miriam and Farnadi, Golnoosh and Fioretto, Ferdinando and Kusner, Matt and Schrouff, Jessica}, volume = {214}, series = {Proceedings of Machine Learning Research}, month = {03 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v214/juarez23a/juarez23a.pdf}, url = {https://proceedings.mlr.press/v214/juarez23a.html}, abstract = {As in traditional machine learning models, models trained with federated learning may exhibit disparate performance across demographic groups. Model holders must identify these disparities to mitigate undue harm to the groups. However, measuring a model’s performance in a group requires access to information about group membership which, for privacy reasons, often has limited availability. We propose novel locally differentially private mechanisms to measure differences in performance across groups while protecting the privacy of group membership. To analyze the effectiveness of the mechanisms, we bound their error in estimating a disparity when optimized for a given privacy budget. Our results show that the error rapidly decreases for realistic numbers of participating clients, demonstrating that, contrary to what prior work suggested, protecting privacy is not necessarily in conflict with identifying performance disparities of federated models.} }
Endnote
%0 Conference Paper %T “You Can’t Fix What You Can’t Measure”: Privately Measuring Demographic Performance Disparities in Federated Learning %A Marc Juarez %A Aleksandra Korolova %B Proceedings of the Workshop on Algorithmic Fairness through the Lens of Causality and Privacy %C Proceedings of Machine Learning Research %D 2023 %E Awa Dieng %E Miriam Rateike %E Golnoosh Farnadi %E Ferdinando Fioretto %E Matt Kusner %E Jessica Schrouff %F pmlr-v214-juarez23a %I PMLR %P 67--85 %U https://proceedings.mlr.press/v214/juarez23a.html %V 214 %X As in traditional machine learning models, models trained with federated learning may exhibit disparate performance across demographic groups. Model holders must identify these disparities to mitigate undue harm to the groups. However, measuring a model’s performance in a group requires access to information about group membership which, for privacy reasons, often has limited availability. We propose novel locally differentially private mechanisms to measure differences in performance across groups while protecting the privacy of group membership. To analyze the effectiveness of the mechanisms, we bound their error in estimating a disparity when optimized for a given privacy budget. Our results show that the error rapidly decreases for realistic numbers of participating clients, demonstrating that, contrary to what prior work suggested, protecting privacy is not necessarily in conflict with identifying performance disparities of federated models.
APA
Juarez, M. & Korolova, A.. (2023). “You Can’t Fix What You Can’t Measure”: Privately Measuring Demographic Performance Disparities in Federated Learning. Proceedings of the Workshop on Algorithmic Fairness through the Lens of Causality and Privacy, in Proceedings of Machine Learning Research 214:67-85 Available from https://proceedings.mlr.press/v214/juarez23a.html.

Related Material