arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1707.07708stat.MLcs.LG
收藏

每实例差异隐私

Per-instance Differential Privacy

UC Santa Barbara(加州大学圣巴巴拉分校);Carnegie Mellon University(卡内基梅隆大学)

Yu-Xiang Wang

摘要:

我们考虑一个细化的差分隐私-每实例差分隐私(pDP),它捕获一个特定的个人相对于一个固定的数据集的隐私。我们表明,这是一个严格的推广标准DP和继承其所有可取的性质,例如,合成,对边信息的不变性和对后处理的封闭性,除了它们都分别适用于每个实例。当数据来自一个分布,我们表明,每实例DP意味着泛化。此外,我们提供了明确的计算每实例DP的输出扰动一类光滑学习问题。结果揭示了一个有趣而直观的事实,即如果一个人对数据集具有较小的“杠杆分数”,并且如果他/她可以使用留一数据集更准确地预测,则他/她具有更强的隐私。我们的模拟显示了几个数量级更有利的隐私和效用权衡时,我们只考虑数据集中的用户的隐私。在差分私有线性回归的案例研究中,提供了一种新的单后验样本(OPS)估计量的分析,并表明当数据集条件良好时,它为任何目标个体提供$(ε,δ)$ -pDP,并将精确下限匹配到$1+\tilde{O}(n^{-1}ε^{-2})$乘法因子。我们还演示了如何使用“pDP到DP转换”步骤来设计AdaOPS,该AdaOPS使用自适应正则化来实现与$(ε,δ)$ -DP相同的结果。

英文摘要:

We consider a refinement of differential privacy --- per instance differential privacy (pDP), which captures the privacy of a specific individual with respect to a fixed data set. We show that this is a strict generalization of the standard DP and inherits all its desirable properties, e.g., composition, invariance to side information and closedness to postprocessing, except that they all hold for every instance separately. When the data is drawn from a distribution, we show that per-instance DP implies generalization. Moreover, we provide explicit calculations of the per-instance DP for the output perturbation on a class of smooth learning problems. The result reveals an interesting and intuitive fact that an individual has stronger privacy if he/she has small "leverage score" with respect to the data set and if he/she can be predicted more accurately using the leave-one-out data set. Our simulation shows several orders-of-magnitude more favorable privacy and utility trade-off when we consider the privacy of only the users in the data set. In a case study on differentially private linear regression, provide a novel analysis of the One-Posterior-Sample (OPS) estimator and show that when the data set is well-conditioned it provides $(ε,δ)$-pDP for any target individuals and matches the exact lower bound up to a $1+\tilde{O}(n^{-1}ε^{-2})$ multiplicative factor. We also demonstrate how we can use a "pDP to DP conversion" step to design AdaOPS which uses adaptive regularization to achieve the same results with $(ε,δ)$-DP.

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。

↑