Differentially Private Nonparametric Hypothesis Testing

Differentially Private Nonparametric Hypothesis Testing
复制标题

差分隐私非参数假设检验

DOI:
10.1145/3319535.3339821
复制
发表时间:
2019
期刊:
Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security (CCS
影响因子:
--
通讯作者:
Groce, Adam
Groce, Adam
中科院分区:
--
文献类型:
--
作者:
Couch, Simon;Kazan, Zeki;Shi, Kaiyan;Bray, Andrew;Groce, Adam

文献摘要

参考文献

被引文献

相似文献

假设检验是数据挖掘的重要统计工具,也是许多领域科学研究的主力。在这里,我们研究一个分类变量和一个连续变量之间的独立性差异私人测试。我们以传统的非参数检验为起点,它不需要分布假设(例如,正态性(normality)。我们提出了私人类似物的Kruskal-Wallis,Mann-Whitney和Wilcoxon符号秩检验,以及参数单样本t检验。这些测试使用专门为私人环境开发的新测试统计数据。我们将我们的测试与先前的工作进行比较,包括参数和非参数测试。我们发现,在所有情况下,我们的新的非参数检验取得了很大的改善统计能力,即使参数检验的假设得到满足。
Hypothesis tests are a crucial statistical tool for data mining and are the workhorse of scientific research in many fields. Here we study differentially private tests of independence between a categorical and a continuous variable. We take as our starting point traditional nonparametric tests, which require no distributional assumption (e.g., normality) about the data distribution. We present private analogues of the Kruskal-Wallis, Mann-Whitney, and Wilcoxon signed-rank tests, as well as the parametric one-sample t-test. These tests use novel test statistics developed specifically for the private setting. We compare our tests to prior work, both on parametric and nonparametric tests. We find that in all cases our new nonparametric tests achieve large improvements in statistical power, even when the assumptions of parametric tests are met.
配对样本数据的差异私有显着性测试
DOI: 10.1137/1.9781611974348.18
发表时间: 2016
影响因子: 3.6
作者:
Christine Task;Chris Clifton
通讯作者: Chris Clifton
重新审视分类数据的差分隐私假设检验
DOI: --
发表时间: 2015
期刊:
影响因子: --
作者:
Yue Wang;Jaewoo Lee;Daniel Kifer
通讯作者: Daniel Kifer
差异私人方差分析测试
DOI: --
发表时间: 2017
期刊: International Conference on Data Intelligence and Security
影响因子: --
作者:
Zachary Campbell;Andrew Bray;Anna M. Ritz;Adam Groce
通讯作者: Adam Groce
DOI: 10.1080/01621459.1973.10481460
发表时间: 1973-12
影响因子: 3.7
作者:
William Jay Conover
通讯作者: William Jay Conover
DOI: 10.2478/popets-2019-0049
发表时间: 2019
影响因子: --
作者:
Swanberg, Marika;Globus-Harris, Ira;Griffith, Iris;Ritz, Anna;Groce, Adam;Bray, Andrew
通讯作者: Bray, Andrew