REVIEW 2 cited by
Canary in a Coalmine: Better Membership Inference with Ensembled Adversarial Queries
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
As industrial applications are increasingly automated by machine learning models, enforcing personal data ownership and intellectual property rights requires tracing training data back to their rightful owners. Membership inference algorithms approach this problem by using statistical techniques to discern whether a target sample was included in a model's training set. However, existing methods only utilize the unaltered target sample or simple augmentations of the target to compute statistics. Such a sparse sampling of the model's behavior carries little information, leading to poor inference capabilities. In this work, we use adversarial tools to directly optimize for queries that are discriminative and diverse. Our improvements achieve significantly more accurate membership inference than existing methods, especially in offline scenarios and in the low false-positive regime which is critical in legal settings. Code is available at https://github.com/YuxinWenRick/canary-in-a-coalmine.
Forward citations
Cited by 2 Pith papers
-
Unveiling Client Privacy Leakage from Public Dataset Usage in Federated Distillation
A server-side attacker can infer label distributions and training-set membership of clients in public-dataset-assisted federated distillation using only black-box logit access.
-
Has My System Prompt Been Used? Large Language Model Prompt Membership Inference
A permutation test on BERT embeddings of LLM outputs can detect, with statistical significance, when response distributions differ because a chat service uses a different system prompt than a candidate prompt.
Discussion (0). Continue with ORCID to comment.