User-Level Differentially Private Mean Estimation for Real-World Datasets

User-Level Differentially Private Mean Estimation for Real-World Datasets

Authors: V. A. Rameshwar, A. Tandon, and A. Sharma

In this work, we provide rigorous theoretical justifications for the performance trends of well-known clipping-based algorithms on real-world ITMS and i.i.d. synthetic datasets. An important contribution of this work is the formalization and explicit computation of the “worst-case estimation error” incurred by a canonical user-level differentially private algorithm for mean estimation, which clips the number of contributions of users in an attempt to increase the accuracy of reconstruction.

Journal/Conference

2024 IEEE International Symposium on Information Theory (ISIT) – Workshop on Information-Theoretic Methods for Trustworthy Machine Learning (IT-TML)