arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.09341math.STcs.ITmath.ITmath.PRstat.TH
收藏

置换混合物的近似独立性

Approximate independence of permutation mixtures

Yanjun Han, Jonathan Niles-Weed

AI总结:

本文提出控制χ²散度的新方法,证明了高维可交换混合分布与其独立同分布对应物的统计距离界限,并得出新的de Finetti风格定理及差分隐私等统计结果。

摘要:

我们证明了高维可交换混合分布(我们称之为置换混合物)与其独立同分布对应物之间统计距离的界限。我们的结果基于一种控制可交换混合物之间χ²散度的新颖方法,该方法比现有的矩或累积量方法更紧。在技术层面上,我们证明中的一个关键创新是对求和为零的变量的初等对称多项式的新麦克劳林型不等式,以及双随机半正定矩阵永久式的上界。作为推论,我们得到了一个新的de Finetti风格定理(采用Diaconis和Freedman, 1987的语言),以及若干新的统计结果,包括带有高斯噪声的“洗牌隐私模型”的差分隐私保证,以及复合决策问题中经验贝叶斯过程的改进通用一致性保证。

英文摘要:

We prove bounds on statistical distances between high-dimensional exchangeable mixture distributions (which we call \emph{permutation mixtures}) and their i.i.d. counterparts. Our results are based on a novel method for controlling $χ^2$ divergences between exchangeable mixtures, which is tighter than the existing methods of moments or cumulants. At a technical level, a key innovation in our proofs is a new Maclaurin-type inequality for elementary symmetric polynomials of variables that sum to zero and an upper bound on permanents of doubly-stochastic positive semidefinite matrices. We obtain as a corollary a new de Finetti-style theorem (in the language of Diaconis and Freedman, 1987), as well as several new statistical results, including a differential privacy guarantee for the ``shuffled privacy model'' with Gaussian noise and improved generic consistency guarantees for empirical Bayes procedures in compound decision problems.

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。

↑