AI Paper Atlas

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

Fast Adversarial Training with Smooth Convergence

Mengnan Zhao, Lihe Zhang, Yuqiu Kong, Baocai Yin

ICCV 2023 · Adversarial robustness and security › Certified and empirical robustness · approach: Adversarial training refinements · 16 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 batch-based ConvergeSmooth to FGSM-MEPFGSM-MEP · CIFAR-10 (eps=16/255)AutoAttack accuracy: 18.98 -> 23.68Improved
add batch-based ConvergeSmooth to FGSM-RSFGSM-RS · CIFAR-10 (eps=16/255)AutoAttack accuracy: 14.84 -> 19.43Improved
add batch-based ConvergeSmooth to FGSM-MEPFGSM-MEP · CIFAR-10 (eps=16/255)PGD-50 accuracy: 26.56 -> 32.95Improved
add weight centralization to FGSM-RSFGSM-RS · CIFAR-10 (eps=16/255)AutoAttack accuracy: 14.84 -> 17.14Improved
add example-based ConvergeSmooth to FGSM-MEPFGSM-MEP · CIFAR-10 (eps=16/255)AutoAttack accuracy: 18.98 -> 17.74Hurt
add batch-based ConvergeSmooth to FGSM-BP on Tiny ImageNetFGSM-BP · Tiny ImageNet (PreActResNet18)AutoAttack accuracy: 3.62 -> 5.95Improved
add ConvergeSmooth to FGSM-MEP with WideResNet34-10FGSM-MEP · CIFAR-10 (WideResNet34-10)AutoAttack accuracy: 15.43 -> 24.27Improved
increase gamma_max above 0.09 (ablation)B-RS · CIFAR-100 (ResNet18)stabilityHurt
use dynamic convergence stride (gamma_max/gamma_min = 1.5) instead of static (ablation)B-RS · CIFAR-100 (ResNet18)robustnessImproved
increase loss coefficient w1 from 0 to 0.9 (ablation)B-RS · CIFAR-100 (ResNet18, eps=12/255)AutoAttack accuracy: 15.28 -> 13.53Hurt
increase w2 from 0.5 to 1.0 (ablation)B-RS · CIFAR-100 (ResNet18, eps=12/255)AutoAttack accuracy: 12.23 -> 15.28Improved
set weight centralization coefficient w3 to 0 vs 0.1 (ablation)weight centralization · CIFAR-10 / CIFAR-100stabilityImproved

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