[论文] Propose-Filter-Release: Scaling Noise to Typical Sensitivity via Priva...
研究领域: ML 作者: Moshe Shenfeld, Eliad Tsfadia, Katrina Ligett 发布时间: 2025-10-28 arXiv: 2510.25919
论文概要
研究领域: ML 作者: Moshe Shenfeld, Eliad Tsfadia, Katrina Ligett 发布时间: 2025-10-28 arXiv: 2510.25919
中文摘要
一个估计器在典型数据上可以既准确又稳定,但对少数可能的记录敏感。差分隐私机制通常将噪声缩放到单条记录的最坏情况影响,这让这些记录即使在剩余数据允许更小尺度的情况下也决定了噪声水平。如果先移除它们就能恢复更小的尺度,但移除过程本身必须是隐私的。我们提出 propose-filter-release (PFR),通过私密地过滤其输入,将一个保护添加或移除典型记录(即适合剩余数据的记录)的机制,转变为在每种输入上都满足标准差分隐私的机制。使用我们新颖的过滤步骤,PFR只需要一个凹的友好度分数(衡量一条记录与数据集的契合程度)和一个友好差分隐私的释放机制(即当改变的记录适合剩余数据时具有隐私性)。对于每一个这样的分数和释放机制,它在关于记录数量的公共条件下保证差分隐私,因此它是一个通用的隐私原语,可与 propose-test-release 相媲美,而 FriendlyCore 可以被视为其实现之一。我们通过为标量统计量、主子空间估计、用户直方图的KL散度、图的有效电阻和处理规则选择实例化PFR来展示其通用性。
原文摘要
An estimator can be accurate and stable on typical data, yet sensitive to a few possible records. Differentially private mechanisms commonly scale their noise to the worst-case influence of a single record, which lets these records dictate the noise even when the remaining data would allow far less. Removing them first would restore the smaller scale, but the removal must itself be private. We introduce propose-filter-release (PFR), which, by privately filtering its input, turns a mechanism that protects the addition or removal of a typical record, one that fits the remaining data, into one that is differentially private in the standard sense, on every input. Using our novel filtering step, PFR requires only a concave friendliness score, which measures how well a record fits a dataset, and...
*自动采集于 2026-11-01*
#论文 #arXiv #ML #小凯