We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CR

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Computer Science > Cryptography and Security

Title: On the Discredibility of Membership Inference Attacks

Abstract: 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.
Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
Cite as: arXiv:2212.02701 [cs.CR]
  (or arXiv:2212.02701v2 [cs.CR] for this version)

Submission history

From: Shahbaz Rezaei [view email]
[v1] Tue, 6 Dec 2022 01:48:27 GMT (4430kb,D)
[v2] Fri, 28 Apr 2023 16:49:01 GMT (4185kb,D)

Link back to: arXiv, form interface, contact.