AI Paper Atlas

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

A Spectral View of Randomized Smoothing under Common Corruptions: Benchmarking and Improving Certified Robustness

Jiachen Sun, Akshay Mehra, Bhavya Kailkhura, Pin-Yu Chen, Dan Hendrycks, Jihun Hamm, Z. Morley Mao

ECCV 2022 · Adversarial robustness and security › Certified and empirical robustness · approach: Randomized smoothing certification · 11 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 hierarchical consistency regularization (HCR) in place of JSD to FourierMix (ablation)FourierMix (ResNet-110) · CIFAR-10-CmACR: 0.444 -> 0.460Improved
add HCR in place of JSD to FourierMix (ablation)FourierMix (ResNet-110) · CIFAR-10-C-barmACR: 0.397 -> 0.419Improved
add HCR in place of JSD to FourierMix (ablation)FourierMix (ResNet-110) · CIFAR-100-CmACR: 0.247 -> 0.261Improved
add HCR in place of JSD to FourierMix (ablation)FourierMix (ResNet-50) · ImageNet-CmACR: 0.399 -> 0.397No clear effect
add JSD consistency loss to Gaussian augmentation (ablation)Gaussian augmentation (ResNet-110) · CIFAR-10-CmACR: 0.363 -> 0.439Improved
add JSD consistency loss to Gaussian augmentation (ablation)Gaussian augmentation (ResNet-50) · ImageNet-CmACR: 0.256 -> 0.395Improved
use FourierMix instead of AugMix augmentation, both with HCR (ablation)HCR-regularized training (ResNet-110) · CIFAR-10-CmACR: 0.437 -> 0.460 (5.3% relative gain)Improved
use FourierMix instead of AugMix augmentation, both with HCR (ablation)HCR-regularized training (ResNet-110) · CIFAR-100-CmACR: 0.249 -> 0.261 (4.8% relative gain)Improved
use FourierMix+HCR instead of Gaussian augmentationResNet-110 · CIFAR-10-CmACR: 0.363 -> 0.460 (26.7% relative gain)Improved
use FourierMix+HCR instead of Gaussian augmentationResNet-110 · CIFAR-100-CmACR: 0.169 -> 0.261 (54.4% relative gain)Improved
use FourierMix instead of AugMix, both with JSD (ablation)JSD-regularized training (ResNet-50) · ImageNet-CmACR: 0.391 -> 0.399 (2.1% relative gain on overall mACR vs AugMix+JSD)Improved
add AutoAugment to Gaussian training (ablation)Gaussian augmentation (ResNet-110) · CIFAR-10-C-barmACR: 0.314 -> 0.304Hurt

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