arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.02701cs.CRcs.AI
收藏

论成员推理攻击的不可信性

On the Discredibility of Membership Inference Attacks

University of California(加利福尼亚大学)

Shahbaz Rezaei, Xin Liu

AI总结:

该研究指出成员推理攻击存在高假阳性率问题,会误将成员样本的邻近非成员判为成员,质疑其在司法审计等实际场景的可信度,说明其仅能识别被记忆的子群体,无法定位具体训练样本。

摘要:

随着机器学习模型的广泛应用,研究基于敏感数据训练的模型潜在的数据泄露问题已变得至关重要。近年来,研究者提出了多种成员推理(membership inference, MI)攻击方法,用于判定某个样本是否属于训练集。问题在于这些攻击能否在实际场景中可靠使用。我们发现,MI模型经常将某个成员样本的邻近非成员样本误分类为成员样本。换言之,它们在自身能够识别的精确成员样本所属的子群体上,假阳性率很高。随后我们展示了MI攻击的一个实际应用场景,该问题会在其中产生现实影响:在该场景中,外部审计员(调查人员)利用MI攻击向法官/陪审团证明被审计方非法使用了敏感数据。由于MI攻击在成员子群体上的假阳性率较高,被审计方通过披露MI攻击在这些子群体上的性能表现,对审计员的可信度提出质疑。我们认为,当前的成员推理攻击能够识别出被记忆的子群体,但无法可靠判定该子群体中究竟哪个具体样本被用于训练过程。

英文摘要:

With the wide-spread application of machine learning models, it has become critical to study the potential data leakage of models trained on sensitive data. Recently, various membership inference (MI) attacks are proposed to determine if a sample was part of the training set or not. The question is whether these attacks can be reliably used in practice. We show that MI models frequently misclassify neighboring nonmember samples of a member sample as members. In other words, they have a high false positive rate on the subpopulations of the exact member samples that they can identify. We then showcase a practical application of MI attacks where this issue has a real-world repercussion. Here, MI attacks are used by an external auditor (investigator) to show to a judge/jury that an auditee unlawfully used sensitive data. Due to the high false positive rate of MI attacks on member's subpopulations, auditee challenges the credibility of the auditor by revealing the performance of the MI attacks on these subpopulations. We argue that current membership inference attacks can identify memorized subpopulations, but they cannot reliably identify which exact sample in the subpopulation was used during the training.

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。

↑