AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

CD-UAP: Class Discriminative Universal Adversarial Perturbation

Chaoning Zhang, Philipp Benz, Tooba Imtiaz, In-So Kweon

AAAI 2020 · Adversarial robustness and security › Attacks on text and graphs · approach: Defenses against universal perturbations · 59 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
use CD-UAP with bounded-logit loss for all-class attackImageNet AC-UAP · ImageNetfooling rate: 93.7 vs 83.7 (VGG16); 94.2 vs 80.1 (VGG19); 90.2 vs 84.0 (ResNet152); 85.9 vs 82.7 (Inception-V3)Improved
use half-half targeted/non-targeted batch sampling (ablation)CD-UAP · CIFAR100, classes 0–4AAD_t: 8.00 -> 48.80Improved
use half-half targeted/non-targeted batch sampling (ablation)CD-UAP · CIFAR100, classes 0–4AAD_nt: 1.79 -> 17.31Hurt
remove half-half sampling and compensate with α=19, β=1 (ablation)CD-UAP · CIFAR100, classes 0–4AAD_t: 48.80 -> 48.20Hurt
remove half-half sampling and compensate with α=19, β=1 (ablation)CD-UAP · CIFAR100, classes 0–4AAD_nt: 17.31 -> 19.43Hurt
train using only targeted-class samples rather than both groups (ablation)CD-UAP · CIFAR100, classes 0–4AAD_nt: 17.31 -> 53.67Hurt
use bounded-logit loss rather than unbounded-logit loss for targeted samples (ablation)CD-UAP · CIFAR100AAD_nt: 46.02 -> 7.16Improved
use cross-entropy loss for non-targeted samples instead of bounded-logit loss (ablation)CD-UAP with targeted bounded-logit loss · CIFAR100AAD_nt: 17.79 -> 17.31Improved
target two superclasses simultaneously instead of one (ablation)CD-UAP · CIFAR100 superclassesΔ_AADHurt
transfer a perturbation between networks in the same architecture familyCD-UAP · ImageNet transferabilityΔ_AAD: 51.65* vs 25.92 (VGG16 to VGG19); 54.81* vs 30.92 (VGG19 to VGG16)Mixed

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