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 2605.22952cs.DB
收藏

通过差分隐私下的依赖量化测量数据库不公平性

Measuring Database Unfairness via Dependency Quantification Under Differential Privacy

Hebrew University(希伯来大学)

Mariia Vologdin, Yuchao Tao, Amir Gilad

AI总结:

提出一个形式化框架,在差分隐私下通过三种互补度量(基于互信息、数据修复和top-k元组贡献)量化数据不公平性,并设计隐私保护算法进行有效近似。

摘要:

差分隐私已成为保护敏感数据的事实标准,提供了强有力的保证,即发布的统计信息或模型仅泄露关于任何个体的有限信息。然而,隐私噪声和受限的数据访问使得评估私有数据集的公平性和可靠性变得越来越困难。在本文中,我们提出了一个在差分隐私下量化数据不公平性的形式化框架。我们基于先前工作确定了不公平性度量的三个核心需求:正性、单调性和差分隐私可计算性。我们通过三种互补度量进一步实例化它们:(1)基于互信息的度量,使用适用于差分隐私的总变差距离代理;(2)基于数据修复的度量,通过归约到加权MaxSAT进行近似;(3)top-k元组贡献度量,隔离公平性违规中最有影响力的记录。我们设计了隐私保护算法,并分析了它们的准确性、效率和隐私成本。在多个真实世界数据集上的大量实验表明,我们提出的度量能够忠实地近似其非隐私对应物,在隐私约束下有效量化偏差,并为数据管理提供见解。

英文摘要:

Differential privacy (DP) has become the de facto standard for protecting sensitive data, providing strong guarantees that published statistics or models reveal limited information about any individual. However, privacy noise and restricted data access make it increasingly difficult to assess the fairness and reliability of private datasets. In this paper, we propose a formal framework for quantifying data unfairness under DP. We identify three core desiderata for unfairness measures based on previous work: positivity, monotonicity, and DP computability. We further instantiate them through three complementary measures: (1) a mutual information-based measure with a total variation distance proxy suitable for DP, (2) a data repair-based measure approximated via a reduction to weighted MaxSAT, and (3) a top-$k$ tuple contribution measure that isolates the most influential records in fairness violations. We design privacy-preserving algorithms and analyze their sensitivity, accuracy, and efficiency. Extensive experiments on multiple real-world datasets demonstrate that our proposed measures faithfully approximate their non-private counterparts, effectively quantify bias under privacy constraints, and provide insights for data management.

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。

↑