AI Paper Atlas

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

ScaleCert: Scalable Certified Defense against Adversarial Patches with Sparse Superficial Layers

Husheng Han, Kaidi Xu, Xing Hu, Xiaobing Chen, Ling Liang, Zidong Du, Qi Guo, Yanzhi Wang, Yunji Chen

NeurIPS 2021 · Adversarial robustness and security › Certified and empirical robustness · approach: Provable defenses and their limits · 13 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
use SIN-based sparse pruning and certified detectionScaleCert · ImageNet, 1% patchcertified accuracy: 36.3 -> 60.4Improved
use SIN-based sparse pruning and certified detectionScaleCert · ImageNet, 2% patchcertified accuracy: 33.9 -> 55.4Improved
use SIN-based sparse pruning and certified detectionScaleCert · ImageNet, 3% patchcertified accuracy: 31.1 -> 52.8Improved
use ScaleCert instead of MRcertified defense on CIFAR-10 · CIFAR-10, 0.4% patchcertified accuracy: 82.5 -> 81.0Hurt
use ScaleCert instead of MRcertified defense on CIFAR-10 · CIFAR-10, 2.4% patchcertified accuracy: 78.1 -> 75.3Hurt
reduce top-ranking rate from 20% to 10% (ablation)ScaleCert · ImageNet, 1% patchcertified accuracy: 60.4 -> 56.0Hurt
reduce top-ranking rate from 20% to 15% (ablation)ScaleCert · ImageNet, 1% patchcertified accuracy: 60.4 -> 58.4Hurt
reduce top-ranking rate from 20% to 10% (ablation)ScaleCert · ImageNet, 2% patchcertified accuracy: 53.3 -> 51.8Hurt
reduce top-ranking rate from 20% to 15% (ablation)ScaleCert · ImageNet, 2% patchcertified accuracy: 53.3 -> 54.9Improved
reduce top-ranking rate from 20% to 10% (ablation)ScaleCert · ImageNet, 3% patchcertified accuracy: 51.6 -> 48.6Hurt
reduce top-ranking rate from 20% to 15% (ablation)ScaleCert · ImageNet, 3% patchcertified accuracy: 51.6 -> 49.6Hurt
set overlap ratio threshold to 0.3 rather than 0.5 (ablation)ScaleCert · ImageNet, 2% patchcertified accuracy: 55.2 -> 56.6Improved

Later papers that compared with this method

1 later accepted paper reports ScaleCert as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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