Concurrent Composition Theorems for Differential Privacy

被引:1
作者
Vadhan, Salil [1 ]
Zhang, Wanrong [1 ]
机构
[1] Harvard Univ, Cambridge, MA 02138 USA
来源
PROCEEDINGS OF THE 55TH ANNUAL ACM SYMPOSIUM ON THEORY OF COMPUTING, STOC 2023 | 2023年
关键词
Differential Privacy; Interactive Mechanisms; Concurrent Composition; NOISE;
D O I
10.1145/3564246.3585241
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
We study the concurrent composition properties of interactive differentially private mechanisms, whereby an adversary can arbitrarily interleave its queries to the different mechanisms. We prove that all composition theorems for non-interactive differentially private mechanisms extend to the concurrent composition of interactive differentially private mechanisms, whenever differential privacy is measured using the hypothesis testing framework of f-DP, which captures standard (epsilon, delta)-DP as a special case. We prove the concurrent composition theorem by showing that every interactive f-DP mechanism can be simulated by interactive post-processing of a non-interactive f-DP mechanism. In concurrent and independent work, Lyu (NeurIPS '22) proves a similar result to ours for (epsilon,delta)-DP, as well as a concurrent composition theorem for Renyi DP. We also provide a simple proof of Lyu's concurrent composition theorem for Renyi DP. Lyu leaves the general case of f-DP as an open problem, which we solve in this paper.
引用
收藏
页码:507 / 519
页数:13
相关论文
共 21 条
[21]   A Statistical Framework for Differential Privacy [J].
Wasserman, Larry ;
Zhou, Shuheng .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2010, 105 (489) :375-389