arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.01559cs.CRcs.LG
收藏

平滑差分隐私

Smoothed Differential Privacy

Google(谷歌);UC Santa Barbara(加州大学圣巴巴拉分校);Rensselaer Polytechnic Institute(伦斯勒理工学院)

Ao Liu, Yu-Xiang Wang, Lirong Xia

AI总结:

针对传统差分隐私需加性噪声、不适用于部分现实场景的问题,提出基于平滑分析的平滑差分隐私概念,证明其相关性质并给出高效参数计算算法,验证了离散机制在现实场景下无加性噪声也可具备隐私性。

摘要:

差分隐私(DP)是一种基于最坏情况分析、被广泛接受且应用广泛的隐私概念。通常,DP会将大多数无加性噪声的机制归类为非隐私的(Dwork等人,2014),因此人们会添加加性噪声来提升隐私性(以实现DP)。但在许多现实应用中,添加加性噪声是不可取的(Bagdasaryan等人,2019),有时甚至是被禁止的(Liu等人,2020)。\n本文中,我们遵循著名的平滑分析(Spielman & Teng,2004年5月)背后的最坏平均情况思路,提出了DP的一种自然扩展。我们提出的概念——平滑DP(smoothed DP),能够在现实场景下有效衡量无加性噪声的机制的隐私泄露情况。我们证明,任何带采样过程的离散机制,其隐私性都优于DP的预测结果;而许多带采样过程的连续机制在平滑DP下仍不具备隐私性。此外,我们证明了平滑DP的若干优良性质,包括组合性、后处理鲁棒性和分布约简性。基于这些性质,我们提出了一种计算平滑DP隐私参数的高效算法。实验方面,我们验证了根据平滑DP,离散采样机制在现实选举场景中具备隐私性,且部分离散神经网络无需添加任何加性噪声即可具备隐私性。我们认为这些结果为超越最坏情况分析的现实隐私衡量标准奠定了理论基础。

英文摘要:

Differential privacy (DP) is a widely-accepted and widely-applied notion of privacy based on worst-case analysis. Often, DP classifies most mechanisms without additive noise as non-private (Dwork et al., 2014). Thus, additive noises are added to improve privacy (to achieve DP). However, in many real-world applications, adding additive noise is undesirable (Bagdasaryan et al., 2019) and sometimes prohibited (Liu et al., 2020). In this paper, we propose a natural extension of DP following the worst average-case idea behind the celebrated smoothed analysis (Spielman & Teng, May 2004). Our notion, smoothed DP, can effectively measure the privacy leakage of mechanisms without additive noises under realistic settings. We prove that any discrete mechanism with sampling procedures is more private than what DP predicts, while many continuous mechanisms with sampling procedures are still non-private under smoothed DP. In addition, we prove several desirable properties of smoothed DP, including composition, robustness to post-processing, and distribution reduction. Based on those properties, we propose an efficient algorithm to calculate the privacy parameters for smoothed DP. Experimentally, we verify that, according to smoothed DP, the discrete sampling mechanisms are private in real-world elections, and some discrete neural networks can be private without adding any additive noise. We believe that these results contribute to the theoretical foundation of realistic privacy measures beyond worst-case analysis.

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。

↑