Improving the Variance of Differentially Private Randomized Experiments through Clustering

Adel Javanmard, Vahab Mirrokni, Jean Pouget-Abadie
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:26946-26980, 2025.

Abstract

Estimating causal effects from randomized experiments is only possible if participants are willing to disclose their potentially sensitive responses. Differential privacy, a widely used framework for ensuring an algorithm’s privacy guarantees, can encourage participants to share their responses without the risk of de-anonymization. However, many mechanisms achieve differential privacy by adding noise to the original dataset, which reduces the precision of causal effect estimation. This introduces a fundamental trade-off between privacy and variance when performing causal analyses on differentially private data. In this work, we propose a new differentially private mechanism, Cluster-DP, which leverages a given cluster structure in the data to improve the privacy-variance trade-off. While our results apply to any clustering, we demonstrate that selecting higher-quality clusters—according to a quality metric we introduce—can decrease the variance penalty without compromising privacy guarantees. Finally, we evaluate the theoretical and empirical performance of our Cluster-DP algorithm on both real and simulated data, comparing it to common baselines, including two special cases of our algorithm: its unclustered version and a uniform-prior version.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-javanmard25a, title = {Improving the Variance of Differentially Private Randomized Experiments through Clustering}, author = {Javanmard, Adel and Mirrokni, Vahab and Pouget-Abadie, Jean}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {26946--26980}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/javanmard25a/javanmard25a.pdf}, url = {https://proceedings.mlr.press/v267/javanmard25a.html}, abstract = {Estimating causal effects from randomized experiments is only possible if participants are willing to disclose their potentially sensitive responses. Differential privacy, a widely used framework for ensuring an algorithm’s privacy guarantees, can encourage participants to share their responses without the risk of de-anonymization. However, many mechanisms achieve differential privacy by adding noise to the original dataset, which reduces the precision of causal effect estimation. This introduces a fundamental trade-off between privacy and variance when performing causal analyses on differentially private data. In this work, we propose a new differentially private mechanism, Cluster-DP, which leverages a given cluster structure in the data to improve the privacy-variance trade-off. While our results apply to any clustering, we demonstrate that selecting higher-quality clusters—according to a quality metric we introduce—can decrease the variance penalty without compromising privacy guarantees. Finally, we evaluate the theoretical and empirical performance of our Cluster-DP algorithm on both real and simulated data, comparing it to common baselines, including two special cases of our algorithm: its unclustered version and a uniform-prior version.} }
Endnote
%0 Conference Paper %T Improving the Variance of Differentially Private Randomized Experiments through Clustering %A Adel Javanmard %A Vahab Mirrokni %A Jean Pouget-Abadie %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-javanmard25a %I PMLR %P 26946--26980 %U https://proceedings.mlr.press/v267/javanmard25a.html %V 267 %X Estimating causal effects from randomized experiments is only possible if participants are willing to disclose their potentially sensitive responses. Differential privacy, a widely used framework for ensuring an algorithm’s privacy guarantees, can encourage participants to share their responses without the risk of de-anonymization. However, many mechanisms achieve differential privacy by adding noise to the original dataset, which reduces the precision of causal effect estimation. This introduces a fundamental trade-off between privacy and variance when performing causal analyses on differentially private data. In this work, we propose a new differentially private mechanism, Cluster-DP, which leverages a given cluster structure in the data to improve the privacy-variance trade-off. While our results apply to any clustering, we demonstrate that selecting higher-quality clusters—according to a quality metric we introduce—can decrease the variance penalty without compromising privacy guarantees. Finally, we evaluate the theoretical and empirical performance of our Cluster-DP algorithm on both real and simulated data, comparing it to common baselines, including two special cases of our algorithm: its unclustered version and a uniform-prior version.
APA
Javanmard, A., Mirrokni, V. & Pouget-Abadie, J.. (2025). Improving the Variance of Differentially Private Randomized Experiments through Clustering. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:26946-26980 Available from https://proceedings.mlr.press/v267/javanmard25a.html.

Related Material