AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Robustness and Generalization via Generative Adversarial Training

Omid Poursaeed, Tianxing Jiang, Harry Yang, Serge Belongie, Ser-Nam Lim

ICCV 2021 · Adversarial robustness and security › Certified and empirical robustness · approach: Robust overfitting and regularization · 29 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
adversarial training with generative (GAT) unrestricted examplesInception-v3 · LSUN clean testaccuracy: 88.9% -> 89.5%Improved
adversarial training with GAT examplesInception-v3 · LSUN adversarial testaccuracy: 0.0% -> 78.4%Improved
adversarial training with GAT examplesgender classifier · CelebA-HQ clean testaccuracy: 95.7% -> 96.2%Improved
adversarial training with GAT examplesDeepLab-v2 · segmentation clean testaccuracy: 67.9% -> 69.1%Improved
adversarial training with GAT examplesRetinaNet · detection clean testaverage precision: 39.0% -> 40.2%Improved
adversarial training with GAT examplescars vs cats classifier · ImageNet out-of-domainaccuracy: 91.4% -> 92.0%Improved
adversarial training with GAT examplescars vs cats classifier · LSUN in-domainaccuracy: 94.2% -> 94.7%Improved
train with GAT instead of AT PGD, mean over unseen attacksInception-v3 · LSUN unseen attacksmean accuracy: 13.0% -> 42.1%Improved
train with GAT instead of AT PGD on PGD attackInception-v3 · LSUN PGD attackaccuracy: 56.7% -> 39.4%Hurt
train with GAT instead of PAT, mean over unseen attacksInception-v3 · LSUN unseen attacksmean accuracy: 36.5% -> 42.1%Improved
train with GAT instead of PAT, clean accuracyInception-v3 · LSUN cleanaccuracy: 72.4% -> 89.5%Improved
replace GAT examples with norm-bounded IFGSM-2 examples (ablation)Inception-v3 · LSUN cleanaccuracy: 89.5% -> 88.4%Hurt

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