AI Paper Atlas

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

Towards Robustness of Deep Neural Networks via Regularization

Yao Li, Martin Renqiang Min, Thomas Lee, Wenchao Yu, Erik Kruus, Wei Wang, Cho-Jui Hsieh

ICCV 2021 · Adversarial robustness and security › Certified and empirical robustness · approach: Robust overfitting and regularization · 15 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
add embedding regularization (ER-Classifier vs Madry's Adv)Madry's Adv · CIFAR10 AutoAttackrobust accuracy (%): 35.6 -> 47.2Improved
add embedding regularization (ER-Classifier vs Madry's Adv)Madry's Adv · MNIST AutoAttackrobust accuracy (%): 69.0 -> 79.0Improved
add embedding regularization (ER-Classifier vs Madry's Adv)Madry's Adv · Tiny ImageNet AutoAttackrobust accuracy (%): 9.70 -> 11.1Improved
use ER-Classifier instead of Defense-GANDefense-GAN · MNIST C&Waccuracy (%): 55.0 -> 99.1Improved
add discriminator-based embedding regularization to encoder-classifier (E-CLA) (ablation)E-CLA · MNIST, CIFAR10, Tiny ImageNet PGDrobust accuracyImproved
add discriminator-based embedding regularization to encoder-classifier (E-CLA) (ablation)E-CLA · STL10 PGDrobust accuracyImproved
use Wasserstein discriminator regularization instead of VAE KL regularization (ablation)VAE-CLA · four datasets PGDrobust accuracyImproved
train without min-max robust optimization (ER-Classifier-) on CIFAR10 under strong attack (ablation)ER-Classifier · CIFAR10 PGDrobust accuracyMixed
add ER-Detector combined code (ER2) versus direct output (ER1) vs KD in familiar PGD detectionKD · MNIST PGDACC-COM (%): 29.87 -> 63.81Improved
use ER-Detector instead of KD when trained on PGD, attacked by C&WKD · MNIST cross-attackACC-COM (%): 10.05 -> 100.00Improved
use ER-Detector instead of KD when trained on PGD, attacked by FGSMKD · CIFAR10 cross-attackACC-NADV (%): 97.89 -> 98.19Mixed

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