AI Paper Atlas

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

Gabor Layers Enhance Network Robustness

Juan C. Pérez, Motasem Alfarra, Guillaume Jeanneret, Adel Bibi, Ali Thabet, Bernard Ghanem, Pablo Arbeláez

ECCV 2020 · Adversarial robustness and security › Certified and empirical robustness · approach: Robustness-accuracy trade-off analysis · 31 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
replace first conv layers with Gabor layersVGG16 · CIFAR10 PGD eps=8/255adversarial accuracy: 23.63 -> 30.11Improved
replace first conv layers with Gabor layersVGG16 · SVHN PGD eps=8/255adversarial accuracy: 5.84 -> 14.57Improved
replace first conv layer with Gabor layerLeNet · MNIST PGD eps=0.2adversarial accuracy: 4.39 -> 7.94Improved
replace first conv layer with Gabor layerWideResNet · SVHN PGD eps=8/255adversarial accuracy: 1.02 -> 1.03No clear effect
replace first conv layer with Gabor layerAlexNet · CIFAR100 PGD eps=2/255adversarial accuracy: 15.05 -> 14.77Hurt
replace conv layers with Gabor layersVGG16 · CIFAR100 clean testaccuracy: 67.54 -> 64.49Hurt
replace conv layer with Gabor layerLeNet · MNIST clean testaccuracy: 99.36 -> 99.03Hurt
add Lipschitz regularizer to Gabor-layered networkLeNet · MNIST PGD eps=0.2adversarial accuracy: 7.94 -> 22.69Improved
add Lipschitz regularizer to Gabor-layered networkVGG16 · SVHN PGD eps=8/255adversarial accuracy: 14.57 -> 15.99Improved
add Lipschitz regularizer to Gabor-layered networkAlexNet · CIFAR100 PGD eps=2/255adversarial accuracy: 14.77 -> 14.68No clear effect
add Gabor layers to adversarially trained networkAlexNet · CIFAR10 adv. trainedadversarial accuracy: 19.24 -> 20.26Improved
add Gabor layers to adversarially trained networkAlexNet · CIFAR100 adv. trainedadversarial accuracy: 13.48 -> 10.94Hurt

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