On the Rényi Differential Privacy of the Shuffle Model
On the Rényi Differential Privacy of the Shuffle Model
复制标题
DOI：
10.1145/3460120.3484794
复制
发表时间：
2021-05
期刊：
影响因子：
--
通讯作者：
Antonious M. Girgis;Deepesh Data;S. Diggavi;A. Suresh;P. Kairouz
中科院分区：
文献类型：
--
作者：
Antonious M. Girgis;Deepesh Data;S. Diggavi;A. Suresh;P. Kairouz
本文研究的中心问题是随机随机数发生器在洗牌隐私模型中的Rényi差分隐私（RDP）保证。在混洗模型中，n个客户端中的每个客户端使用本地差分私有（LDP）机制随机化其响应，并且不可信服务器仅接收客户端响应的随机排列（混洗），而不与每个客户端相关联。本文的主要结果是第一个直接RDP界一般离散局部随机化的洗牌隐私模型，我们开发了新的分析技术，推导出我们的结果可能是独立的利益。在应用程序中，这样的RDP保证在我们使用它来组成几个私有交互时最有用。我们数值证明，对于重要的制度，与组合我们的边界产生的隐私保证的改善，由一个因素的8\times $超过国家的最先进的近似差分隐私（DP）保证（与标准组成）洗牌模型。此外，结合泊松子采样，我们的结果导致至少$10\times$改进与标准组合的子采样近似DP。
The central question studied in this paper is Rényi Differential Privacy (RDP) guarantees for general discrete local randomizers in the shuffle privacy model. In the shuffle model, each of the n clients randomizes its response using a local differentially private (LDP) mechanism and the untrusted server only receives a random permutation (shuffle) of the client responses without association to each client. The principal result in this paper is the first direct RDP bounds for general discrete local randomization in the shuffle privacy model, and we develop new analysis techniques for deriving our results which could be of independent interest. In applications, such an RDP guarantee is most useful when we use it for composing several private interactions. We numerically demonstrate that, for important regimes, with composition our bound yields an improvement in privacy guarantee by a factor of $8\times$ over the state-of-the-art approximate Differential Privacy (DP) guarantee (with standard composition) for shuffle models. Moreover, combining with Poisson subsampling, our result leads to at least $10\times$ improvement over subsampled approximate DP with standard composition.