Membership Inference Attacks Against Machine Learning Models¶
Shokri, R., Stronati, M., Song, C., & Shmatikov, V. (2017). Membership Inference Attacks Against Machine Learning Models. Proceedings of the 38th IEEE Symposium on Security and Privacy (S&P), 3-18.
Cited by¶
2 citations across 2 artifacts.
Each citation links to the sentence it supports in the citing article.
Primes¶
- Side Channel Attack
- In machine-learning security it is membership-inference attacks (a model's confidence reveals whether an input was in the training set), model-extraction attacks, and training-data extraction.
This sourceShows a model's confidence distribution leaks training-set membership — a side channel in ML — with output-perturbation defenses.
- In machine-learning security it is membership-inference attacks (a model's confidence reveals whether an input was in the training set), model-extraction attacks, and training-data extraction.
Mechanisms¶
- Membership Inference Probe
- It maps directly onto the differential-privacy guarantee, which is defined as a bound on exactly this membership advantage.
This sourceDifferential privacy is defined precisely to bound how much any single record's inclusion can change outputs, and so directly limits membership advantage.
- It maps directly onto the differential-privacy guarantee, which is defined as a bound on exactly this membership advantage.
Verification¶
Does it exist? Confirmed. This work's DOI resolves to a registered record, which fixes its identity. That is all it fixes.
Does it back the claim? Not recorded. Neither this nor any other of the 2 citations of this work carries a recorded support check.
Was it audited? Yes. A second, independent pass read the citation against the article text and recorded a verdict.
Support is checked per citation rather than per work — the same source can be cited soundly in one article and wrongly in another. Per-citation recording began recently, so a citation with no recorded check is a gap in the record rather than evidence it went unchecked.
See how references were verified.
Registry ID ref:fc51ecf28c6b · see in the full table