arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1710.00892cs.LGcs.AIcs.CR
收藏

后验采样的Rényi差分隐私机制

Rényi Differential Privacy Mechanisms for Posterior Sampling

University of California, San Diego(加利福尼亚大学圣迭戈分校)

Joseph Geumlek, Shuang Song, Kamalika Chaudhuri

摘要:

使用最近提出的隐私Rényi差分隐私(RDP)的定义,我们重新审视固有的隐私释放一个单一的样本从后验分布。我们利用先验分布的影响来减轻单个数据点的影响。特别是,我们专注于从指数族和特定的广义线性模型,如逻辑回归的采样。我们提出了新的RDP机制,以及提供一个新的RDP分析现有的方法,以增加价值的RDP框架。每种方法都能够实现任意的RDP隐私保证,我们提供了实验结果,他们的功效。

英文摘要:

Using a recently proposed privacy definition of Rényi Differential Privacy (RDP), we re-examine the inherent privacy of releasing a single sample from a posterior distribution. We exploit the impact of the prior distribution in mitigating the influence of individual data points. In particular, we focus on sampling from an exponential family and specific generalized linear models, such as logistic regression. We propose novel RDP mechanisms as well as offering a new RDP analysis for an existing method in order to add value to the RDP framework. Each method is capable of achieving arbitrary RDP privacy guarantees, and we offer experimental results of their efficacy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑