AI Paper Atlas

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

Robust Detection of Adversarial Attacks by Modeling the Intrinsic Properties of Deep Neural Networks

Zhihao Zheng, Pengyu Hong

NeurIPS 2018 · Adversarial robustness and security › Black-box and transfer attacks · approach: Robustness benchmarks and defenses · 82 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
model hidden-state distributions with GMM (I-defender) vs Defense-GAN with 10 runs x 50 iterationsdetector · FGSM MNIST eps=0.3detection AUC: 0.993 vs 0.985Mixed
I-defender vs Defense-GAN at small perturbationdetector · FGSM F-MNIST eps=0.1detection AUC: 0.775 -> 0.9302Improved
I-defender vs Defense-GAN at large perturbationdetector · FGSM MNIST eps=0.3detection AUC: 0.999 vs 0.993No clear effect
unsupervised I-defender vs supervised Subnetwork on unseen attack (Iter-l2)detector · CIFAR-10 semi white-boxdetection AUC: 68.980 -> 87.464Improved
I-defender vs supervised Subnetwork on the training attack Iter-linfdetector · CIFAR-10 semi white-boxdetection AUC: 98.660 vs 79.398Hurt
attacker has defense parameters (source DNN) vs unknown parameters (target DNN)I-defender · CIFAR-10 gray-box Iter-linf eps=0.013detection AUC: 83.261 -> 58.967Hurt
larger classifier WRN46-8 instead of WRN34-8I-defender · CIFAR-10 gray-box Iter-linf eps=0.013detection AUC: 85.203 -> 87.549Improved
increase attacker penalty weight alpha from 1e-3 to 1I-defender · CIFAR-10 gray-box target DNN Iter-linf eps=0.013detection AUC: 85.203 -> 83.945No clear effect
FGSM attack on ImageNetI-defender · ImageNet gray-box FGSM eps=0.3detection AUC: 62.685 vs 97.201 (Iter-linf)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