FriendlyCore: Practical Differentially Private Aggregation
FriendlyCore: Practical Differentially Private Aggregation
复制标题
FriendlyCore：实用的差分私有聚合
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Uri Stemmer
中科院分区：
文献类型：
--
作者：
Eliad Tsfadia;E. Cohen;Haim Kaplan;Y. Mansour;Uri Stemmer
用于常见度量聚合任务（如聚类或平均）的差分私有算法，由于其复杂性或精确结果所需的大量数据点，通常具有有限的实用性。我们提出了一个简单实用的工具，$mathsf{FriendlyCore}$，它从一个不受限制的（伪）度量空间中获取一组点${cal D}$作为输入。当${cal D}$具有有效直径$r$时，$mathsf{FriendlyCore}$返回一个“稳定的”子集${cal C} subseteq {cal D}$，它包含所有点，除了可能有几个异常值，并且被{em证明}具有直径$r$。$mathsf{FriendlyCore}$可用于在私下聚合输入之前对其进行预处理，从而可能简化聚合或提高其准确性。令人惊讶的是，$mathsf{FriendlyCore}$是轻量级的，不依赖于维度。我们通过经验证明了它在提高均值估计和聚类任务（如$k$-means和$k$-GMM）的准确性方面的优势，优于定制方法。
Differentially private algorithms for common metric aggregation tasks, such as clustering or averaging, often have limited practicality due to their complexity or to the large number of data points that is required for accurate results. We propose a simple and practical tool, $mathsf{FriendlyCore}$, that takes a set of points ${cal D}$ from an unrestricted (pseudo) metric space as input. When ${cal D}$ has effective diameter $r$, $mathsf{FriendlyCore}$ returns a"stable"subset ${cal C} subseteq {cal D}$ that includes all points, except possibly few outliers, and is {em certified} to have diameter $r$. $mathsf{FriendlyCore}$ can be used to preprocess the input before privately aggregating it, potentially simplifying the aggregation or boosting its accuracy. Surprisingly, $mathsf{FriendlyCore}$ is light-weight with no dependence on the dimension. We empirically demonstrate its advantages in boosting the accuracy of mean estimation and clustering tasks such as $k$-means and $k$-GMM, outperforming tailored methods.
DOI：
--
发表时间：
2018-05
期刊：
--
影响因子：
--
作者：
Gautam Kamath;Jerry Li;Vikrant Singhal;Jonathan Ullman
通讯作者：
Gautam Kamath;Jerry Li;Vikrant Singhal;Jonathan Ullman
DOI：
--
发表时间：
2021-12
期刊：
--
影响因子：
--
作者：
Pravesh Kothari;Pasin Manurangsi;A. Velingker
通讯作者：
Pravesh Kothari;Pasin Manurangsi;A. Velingker
影响因子：
6
作者：
Kume S;Nagasu H;Nangaku M;Nishiyama A;Nakamoto H;Kashihara N
通讯作者：
Kashihara N