arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.26257cs.LG
收藏

无需影子模型或保留数据的数据集使用推断

Dataset Usage Inference without Shadow Models or Held-out Data

Warsaw University of Technology(华沙理工大学);NASK National Research Institute(NASK国家研究所);CISPA Helmholtz Center for Information Security(CISPA亥姆霍兹信息安全中心)

Wojciech Łapacz, Stanisław Pawlak, Jan Dubiński, Franziska Boenisch, Adam Dziedzic

AI总结:

提出一种无需影子模型和真实保留数据的数据集使用推断方法,通过生成合成非成员样本并利用混合比例估计来量化训练数据占比。

摘要:

我的数据有多少被用于训练机器学习模型?数据集使用推断(DUI)旨在通过估计数据集对模型训练的贡献比例来回答这个问题。然而,现有的DUI方法依赖于在实践中很少成立的假设:它们需要训练昂贵的影子模型来模仿目标模型,并且假设可以访问已知的训练样本和确认未参与训练的分布内保留集。这些条件使得当前方法对于现代大型模型和真实数据所有权争议不切实际。我们引入了一个实用的DUI框架,消除了这些约束。我们的方法既不需要影子模型,也不需要真实的保留数据。相反,它生成合成的非成员样本,提取多样的成员信号,并将DUI视为混合比例估计问题,以估计候选数据集中有多少份额在训练中被使用。在大型图像生成模型上的实验表明,我们的方法能够可靠地量化数据集使用情况,为数据所有者提供了一种实用工具,以确定他们的数据有多少被用于训练模型。

英文摘要:

How much of my data was used to train a machine learning model? Dataset Usage Inference (DUI) aims to answer this by estimating what fraction of a dataset contributed to a model's training. However, existing DUI methods rely on assumptions that rarely hold in practice: they require training expensive shadow models to imitate the target model, and they assume access to both known training samples and an in-distribution held-out set confirmed to be absent from training. These conditions make current approaches impractical for modern large models and real data ownership disputes. We introduce a practical DUI framework that removes these constraints. Our method requires neither shadow models nor real held-out data. Instead, it generates synthetic non-member samples, extracts diverse membership signals, and casts DUI as a mixture proportion estimation problem to estimate what share of the candidate dataset was used during training. Experiments on large image generative models show that our method reliably quantifies dataset usage, providing a practical tool for data owners to determine how much of their data was used to train a model.

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。

↑