AI Paper Atlas

Research map / Federated learning and privacy / Privacy attacks and secure learning

The Privacy Onion Effect: Memorization is Relative

Nicholas Carlini, Matthew Jagielski, Chiyuan Zhang, Nicolas Papernot, Andreas Terzis, Florian Tramer

NeurIPS 2022 · Federated learning and privacy › Privacy attacks and secure learning · approach: Membership inference and auditing · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
remove the 5,000 most vulnerable examples and retrainCIFAR-10 model, LiRA · CIFAR-10 LiRA, FPR 0.01%true-positive rate on remaining examples: 1.5% -> 0.6% (expected 0.1%)Mixed
remove 5,000 hardest-to-attack or random examples instead of the most vulnerable (ablation)CIFAR-10 model, LiRA · CIFAR-10 LiRAROC curve / attack successNo clear effect
add 5,000 random-label CIFAR-100 outliers after removing CIFAR-10 outliers (ablation)CIFAR-10 model, LiRA · CIFAR-10 LiRAattack success on remaining 45,000No clear effect
deduplicate CIFAR-10 (remove 5,275 duplicates) (ablation)CIFAR-10 LiRA · CIFAR-10 LiRAonion effect presence / attack successMixed
remove 25 highest-PrivInf points per targetsecond-layer target points · CIFAR-10 LiRA targetsmembership inference advantage: +0.22 (56% of the 0.38 increase from removing all 5,000)Improved
remove 25 highest-PrivInf points per targetrandom target points · CIFAR-10 LiRA targetsmembership inference advantage: 0.34 -> 0.42Mixed
remove 25 highest-PrivInf points per targetinitially safe target points · CIFAR-10 LiRA targetsmembership inference advantage: 0.02 -> 0.14Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy