Performative Privacy: When Differential Privacy Maximizes Utility

2026-08-28

AI Summary

研究通过差分隐私机制解决数据泄漏导致用户减少的问题，平衡估计噪声与未来参与度，证明了在长期效用上差分隐私可能优于非私有估计。

Abstract

Privacy-preserving learning is often motivated by the idea that protecting users' data can preserve trust and thus participation, improving utility in the long term. However, this claim has not been formalized so far. In parallel, performative learning provides a framework for studying learning systems whose deployment affects the data they later observe. In this work, we bring these two perspectives together and introduce \emph{performative privacy}, where data leakage reduces future participation. We study a simple model where agents repeatedly contribute data for mean estimation but may leave the system when their data is leaked. Privacy is implemented through differentially private mechanisms, creating a trade-off between estimation noise and future participation. We show, through a theoretical study of the dynamics and numerical experiments, that a finite privacy budget can outperform non-private estimation in the long term when the feedback loop between leakage and participation is sufficiently strong. This provides first evidence that differential privacy can be optimal not only as a protection mechanism, but also from the perspective of long-term utility.

Problem

• differential privacy
• long-term utility
• data leakage
• participation
• performative learning

Innovation

• performative privacy
• differential privacy
• long-term utility