AI Paper Atlas

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

Improving Query Efficiency of Black-box Adversarial Attack

Yang Bai, Yuyuan Zeng, Yong Jiang, Yisen Wang, Shu-Tao Xia, Weiwei Guo

ECCV 2020 · Adversarial robustness and security › Black-box and transfer attacks · approach: Query-efficient black-box attacks · 48 citations (OpenAlex)

Read the paper

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
replace NAttack's per-pixel Gaussian with a Neural Process decoder (NP-Attack-R)NAttack · CIFAR-10 untargetedaverage query count: 183 -> 94Improved
replace NAttack with NP-Attack-RNAttack · MNIST untargetedaverage query count: 4,357 -> 1,226Improved
replace NAttack with NP-Attack-RNAttack · MNIST targetedaverage query count: 5,981 -> 2,693Improved
replace NAttack with NP-Attack-RNAttack · ImageNet untargetedaverage query count: 2,075 -> 867Improved
replace NAttack with NP-Attack-RNAttack · ImageNet targetedaverage query count: 14,229 -> 8,001Improved
replace QL with NP-Attack-RQL · MNIST targetedaverage query count: 2,693 -> 2,693No clear effect
replace NAttack with NP-Attack-R on other architecturesNAttack · CIFAR-10 ResNet18average query count: 311 -> 185Improved
optimize latent z too (NP-Attack-RZ) instead of r only (ablation)NP-Attack-R · MNIST untargetedaverage query count: 1,190 -> 1,460Hurt
optimize z only instead of r only (ablation)NP-Attack-R · CIFAR-10 untargetedaverage query count: 96 -> 150Hurt
increase sample size b from 10 to 50 (ablation)NP-Attack-R · MNIST untargetedattack success rate and query count: 98.01% -> 99.00% ASR, 723 -> 1,203 queriesMixed

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