arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2406.06231math.STcs.CRstat.COstat.TH
收藏

未知样本量的私有化数据的统计推断

Statistical Inference for Privatized Data with Unknown Sample Size

University of Pittsburgh(匹兹堡大学);Florida State University(佛罗里达州立大学);Dartmouth College(达特茅斯学院)

Jordan Awan, Andres Felipe Barrientos, Nianqiao Ju

AI总结:

针对无界差分隐私下样本量也需隐私保护的私有化数据,该研究发展相关理论算法,证明抽样分布与后验分布的收敛性,提出两种MCMC与蒙特卡洛EM算法,并将其应用于线性回归模型和微观数据的分析。

摘要:

我们发展了理论与算法,用于分析无界差分隐私(DP)下的私有化数据,其中甚至样本量也被视为需要隐私保护的敏感量。我们证明,当样本量$n$趋于无穷大时,若用于私有化$n$的噪声速率合适,则无界DP与有界DP下的抽样分布之间的距离趋于零;我们还证明,近似贝叶斯计算(ABC)型后验分布在类似假设下收敛。我们进一步给出隐私预算变化情形下的渐近结果,建立了抽样分布的相似性,并证明无界设置下的最大似然估计(MLE)收敛到有界DP下的MLE。为便于在无界DP下对私有化数据进行有效的有限样本贝叶斯推断,我们提出了一种可逆跳变马尔可夫链蒙特卡洛(MCMC)算法,该算法扩展了Ju等人(2022)提出的数据增广MCMC算法。我们还提出了一种蒙特卡洛期望最大化(Monte Carlo EM)算法,用于从有界和无界DP下的私有化数据中计算MLE。我们将所提方法应用于分析线性回归模型,以及使用狄利克雷分布建模的2019年美国时间使用调查微观数据文件。

英文摘要:

We develop both theory and algorithms to analyze privatized data in unbounded differential privacy (DP), where even the sample size is considered a sensitive quantity that requires privacy protection. We show that the distance between the sampling distributions under unbounded DP and bounded DP goes to zero as the sample size $n$ goes to infinity, provided that the noise used to privatize $n$ is at an appropriate rate; we also establish that Approximate Bayesian Computation (ABC)-type posterior distributions converge under similar assumptions. We further give asymptotic results in regimes where the privacy budgets vary, establishing similarity of sampling distributions as well as showing that the MLE in the unbounded setting converges to the bounded-DP MLE. To facilitate valid, finite-sample Bayesian inference on privatized data under unbounded DP, we propose a reversible jump MCMC algorithm which extends the data augmentation MCMC of Ju et al. (2022). We also propose a Monte Carlo EM algorithm to compute the MLE from privatized data in both bounded and unbounded DP. We apply our methodology to analyze a linear regression model as well as a 2019 American Time Use Survey Microdata File which we model using a Dirichlet distribution.

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。

↑