AI Paper Atlas

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

Sparse Adversarial Attack via Perturbation Factorization

Yanbo Fan, Baoyuan Wu, Tuanhui Li, Yong Zhang, Mingyang Li, Zhifeng Li, Yujiu Yang

ECCV 2020 · Adversarial robustness and security › Certified and empirical robustness · approach: Regularized defenses against attacks · 86 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
jointly optimize perturbation magnitude and binary selection factors (SAPF) instead of C&W-l0 heuristic position selectionsparse targeted attack · CIFAR-10 average casel2-norm: 0.805 -> 0.435Improved
SAPF instead of StrAttacksparse targeted attack · CIFAR-10 average casel1-norm: 9.494 -> 8.513Improved
SAPF instead of C&W-l0sparse targeted attack · ImageNet average casel0-norm: 70885 -> 56218Improved
SAPF instead of StrAttacksparse targeted attack · ImageNet average casel2-norm: 0.974 -> 0.719Improved
SAPF instead of SparseFoolsparse targeted attack · ImageNet worst caseASR (%): 46 -> 100Improved
SparseFool instead of SAPF (lower linf)sparse targeted attack · CIFAR-10 average caselinf-norm: 0.064 -> 0.047Hurt
increase cardinality k (number of perturbed positions) (ablation)SAPF · CIFAR-10 and ImageNetl1/l2/linf-normMixed
add group-wise sparsity (SAPF-GS, lambda2=10 vs 0) (ablation)SAPF · ImageNet examples (350 super-pixel regions)perturbation locality and normsMixed
use variance-based pixel weights instead of uniform weights (SAPF-VI) (ablation)SAPF · ImageNet example (conch to fountain)perturbed positions locationImproved
add brightness (variance-mean weight) to variance weight (ablation)SAPF-VI · ImageNet example (conch to fountain)perturbed positions locationImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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