AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Membership Inference Attacks Are Easier on Difficult Problems

Avital Shafran, Shmuel Peleg, Yedid Hoshen

ICCV 2021 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 20 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
subtract single-sample predictability error from reconstruction error (ablation)Pix2pix · FacadesMIA ROCAUC (%): 93.62% -> 97.59%Improved
subtract predictability error from reconstruction error (ablation)Pix2pixHD · CityscapesMIA ROCAUC (%): 96.04% -> 99.09%Improved
subtract predictability error from reconstruction error (ablation)SPADE · ADE20KMIA ROCAUC (%): 85.31% -> 89.79%Improved
subtract predictability error from reconstruction error (ablation)Pix2pix · Maps2satMIA ROCAUC (%): 84.22% -> 85.65%Improved
subtract predictability error from reconstruction error (ablation)PraNet · PolypMIA ROCAUC (%): 96.03% -> 96.38%Improved
use harder task with more output uncertainty (landmarks to face instead of auto-encoding)reconstruction-based MIA · CelebA tasksMIA ROCAUC (%): 50.74% -> 99.54%Improved
increase number of output pixels used by the attack (1 to 200)reconstruction-based MIA · Pix2pixHD, UperNet, HRNetV2MIA ROCAUC (%)Improved
replace human-supervised difficulty score with self-supervised single-sample predictability errorL_mem · Pix2pix FacadesMIA ROCAUC (%): 93.67% -> 97.59%Improved
use in-distribution multi-image predictability error (100 samples) instead of single sampleL_mem · Maps2satMIA ROCAUC (%): 85.65% vs 92.43% (Pix2pix); 99.42% vs 82.87% (Pix2pixHD)Mixed
replace single-sample predictability with BDD-trained shadow predictabilityL_mem · CityscapesMIA ROCAUC (%): 99.29% -> 66.2% (Pix2pixHD); 82.94% -> 74.43% (Pix2pix)Hurt
use shadow-model classifier attack instead of L_memMIA · Maps2sat Pix2pixMIA ROCAUC (%): 85.65% -> 80.15%Hurt
add reconstruction loss to victim training (vs cGAN loss only) (ablation)reconstruction-based MIA on Pix2pixHD · CityscapesMIA ROCAUC (%): 56.58% -> 96.04%Improved

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