AI Paper Atlas

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

A Hamiltonian Monte Carlo Method for Probabilistic Adversarial Attack and Learning

Hongjun Wang, Guanbin Li, Xiaobai Liu, Liang Lin

TPAMI 2020 · Adversarial robustness and security › Certified and empirical robustness · approach: Regularized defenses against attacks · 33 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 accumulated momentum with sampling (HMCAM) to M-PGDM-PGD · CIFAR10 black-box ResNetXt -> ResNet32_Aattack success rate: 10.11% -> 14.11%Improved
use HMCAM instead of M-PGD for white-box attack on VGG16M-PGD · CIFAR10 white-box VGG16attack success rate: 83.64% -> 93.29%Improved
use HMCAM instead of M-PGD on ResNet32 white-boxM-PGD · CIFAR10 white-box ResNet32attack success rate: 98.93% -> 98.76%No clear effect
AI-FGSM (S=1, accumulated momentum only) vs full HMCAM with sampling S=2 (ablation)AI-FGSM · CIFAR10 single-model black-box ResNet32 -> ResNet32_Aattack success rate: 9.56% -> 10.01%Improved
add accumulated momentum (AI-FGSM) to M-PGD (ablation)M-PGD · CIFAR10 ensemble hold-out ResNet32_Aattack success rate: 21.23% -> 28.31%Improved
add HMCAM to ensemble attack on adversarially trained hold-outPGD · CIFAR10 ensemble hold-out ResNet32_Aattack success rate: 13.88% -> 30.11%Improved
use HMCAM with N=10 iterations versus PGD with N=100 (ablation)PGD · CIFAR10 black-boxattack success rateImproved
vary step size alpha over 0.001 to 0.1 times e^-2 (ablation)HMCAM · CIFAR10 white/black-boxattack success rateNo clear effect
replace PGD-10 adversarial training with CAT on Preact-ResNet18PGD-10 adversarial training · CIFAR10 Preact-ResNet18 PGD-20robust accuracy: 41.58% -> 49.37%Improved
replace PGD-10 adversarial training with CAT on Preact-ResNet18PGD-10 adversarial training · CIFAR10 Preact-ResNet18 cleanclean accuracy: 84.96% -> 81.54%Hurt
replace PGD-10 with CAT on Wide ResNet34PGD-10 adversarial training · CIFAR10 WRN34 PGD-20robust accuracy: 48.4% -> 53.3%Improved
replace PGD-10 with CAT on Wide ResNet34 (training time)PGD-10 adversarial training · CIFAR10 WRN34training time (mins): 2602 -> 672Improved

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