AI Paper Atlas

Research map / Adversarial robustness and security / Black-box and transfer attacks

Adversarial Defense by Restricting the Hidden Space of Deep Neural Networks

Aamir Mustafa, Salman Khan, Munawar Hayat, Roland Goecke, Jianbing Shen, Ling Shao

ICCV 2019 · Adversarial robustness and security › Black-box and transfer attacks · approach: Robustness benchmarks and defenses · 158 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 prototype conformity loss to softmax trainingResNet-110 · CIFAR-10 white-box PGD (eps 0.03)robust accuracy: 0.01 -> 27.2Improved
add prototype conformity loss to softmax trainingResNet-110 · CIFAR-100 white-box PGD (eps 0.03)robust accuracy: 0.17 -> 25.9Improved
add prototype conformity loss to softmax trainingCNN-6 · MNIST white-box PGD (eps 0.3)robust accuracy: 0.0 -> 19.9Improved
add prototype conformity loss to softmax trainingResNet-110 · CIFAR-10 black-box PGD (eps 0.03, VGG-19 source)robust accuracy: 29.1 -> 76.4Improved
add prototype conformity lossResNet-110 · CIFAR-10 cleanaccuracy: 90.80 -> 90.45No clear effect
add PGD adversarial training to PC lossResNet-110 · CIFAR-10 white-box PGD (eps 0.03)robust accuracy: 27.2 -> 46.7Improved
use PGD instead of FGSM adversarial examples for AdvTrainResNet-110 · CIFAR-10 white-box PGD (eps 0.03)robust accuracy: 42.5 -> 46.7Improved
add adversarial training to PC lossResNet-110 · CIFAR-100 cleanaccuracy: 71.90 -> 69.11Hurt
attack the joint PC+CE loss (adaptive white-box)PC loss model, ResNet-110 · CIFAR-10 PGD (eps 8/255)robust accuracy: 27.2 -> 27.09No clear effect
apply L_PC after Layer 5 (ablation)ResNet-110 · CIFAR-10 PGD (eps 0.03), L_PC layer ablationrobust accuracy: 0.01 -> 13.56Improved
apply L_PC after Layers 4 and 5 instead of Layer 5 only (ablation)ResNet-110 · CIFAR-10 PGD (eps 0.03), L_PC layer ablationrobust accuracy: 13.56 -> 20.70Improved
apply L_PC after Layers 3, 4 and 5 instead of 4 and 5 (ablation)ResNet-110 · CIFAR-10 PGD (eps 0.03), L_PC layer ablationrobust accuracy: 20.70 -> 27.23Improved

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