AI Paper Atlas

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

l∞-Robustness and Beyond: Unleashing Efficient Adversarial Training

Hadi M. Dolatabadi, Sarah Erfani, Christopher Leckie

ECCV 2022 · Adversarial robustness and security › Certified and empirical robustness · approach: Adversarial training refinements

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 coreset selection (Craig, 50% data) instead of full trainingTRADES · CIFAR-10 ResNet-18training time (min): 344.29 -> 179.20Improved
adversarial coreset selection (Craig) instead of full trainingTRADES · CIFAR-10 ResNet-18robust accuracy (%): 44.19 -> 41.45Hurt
adversarial coreset selection (GradMatch) instead of full trainingl-inf PGD · CIFAR-10 ResNet-18robust accuracy (%): 41.39 -> 45.23Improved
adversarial coreset selection (GradMatch) instead of full trainingl-inf PGD · CIFAR-10 ResNet-18clean accuracy (%): 83.14 -> 80.67Hurt
adversarial coreset selection (GradMatch, 30% data) instead of full trainingl2-PGD · SVHN ResNet-18training time (min): 389.46 -> 125.53Improved
adversarial coreset selection (GradMatch) instead of full trainingl2-PGD · SVHN ResNet-18robust accuracy (%): 53.02 -> 50.41Hurt
adversarial coreset selection (Craig) instead of full PATFast-LPA · CIFAR-10 ResNet-50training time (min): 1682.94 -> 767.34Improved
adversarial coreset selection (GradMatch) instead of full PATFast-LPA · ImageNet-12 ResNet-50unseen robust accuracy (%): 53.23 -> 53.17No clear effect
add adversarial coreset selection (GradMatch) to Fast Adversarial TrainingFAT · CIFAR-10 ResNet-18training time (min): 31.068 -> 16.419Improved
add adversarial coreset selection (GradMatch) to FATFAT · CIFAR-10 ResNet-18clean accuracy (%): 86.20 -> 82.53Hurt
add adversarial coreset selection to FATFAT · CIFAR-10 ResNet-18robust accuracy (%): 47.54 -> 47.88No clear effect
select coreset by gradient matching instead of random selection (ablation)adversarial GradMatch · CIFAR-10 TRADESrelative robust errorImproved

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