AI Paper Atlas

Research map / Adversarial robustness and security / Black-box and transfer attacks

Simultaneously Optimizing Perturbations and Positions for Black-Box Adversarial Patch Attacks

Xingxing Wei, Ying Guo, Jie Yu, Bo Zhang

TPAMI 2022 · Adversarial robustness and security › Black-box and transfer attacks · approach: Query-efficient black-box attacks · 71 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
optimize position and perturbation simultaneously instead of alternately (LO)adversarial patch attack · Face dodging attack on FaceNetASR: 76.82% -> 96.65%Improved
optimize simultaneously instead of two-step PatchAttackadversarial patch attack · Face impersonation attack on FaceNetASR: 68.95% -> 72.83%Improved
optimize simultaneously instead of two-step PatchAttackadversarial patch attack · Face impersonation attack on CosFace50ASR: 35.56% -> 40.08%Improved
use ensemble surrogate instead of single ArcFace50 surrogate (ablation)our method · Face dodging attack on FaceNetASR: 89.37% -> 96.65%Improved
add weight parameter and step size to position-only learning (ablation)our method · Face dodging attack (average)ASR: 69.80% (position only), further improved by weights and step sizeImproved
initialize ZO-AdaMM with SO patch instead of random initialization (ablation)ZO attack · Face dodging attack on FaceNetASR: 65.79% -> 97.73%Improved
initialize ZO attack with SO patch instead of random initialization (ablation)ZO attack · Face impersonation attack on FaceNetASR: 50.52% -> 87.37%Improved
apply DW preprocessing defenseour attack · Face dodging attack on FaceNetASR: 96.65% -> 81.71%Hurt
apply LGS preprocessing defenseour attack · Face dodging attack on FaceNetASR: drops by 1.01%No clear effect
apply adversarial training to target modelGDPA attack · Face dodging attack on FaceNetASR: down 48.51%Hurt
apply adversarial training to target modelour attack · Face dodging attack on FaceNetASR: down 2.69%No clear effect
translate patch by 5 pixelsour attack · Face dodging attack on FaceNetASR: 96.65% -> 79.24%Hurt

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