AI Paper Atlas

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

Finding Optimal Tangent Points for Reducing Distortions of Hard-label Attacks

Chen Ma, Xiangyu Guo, Li Chen, Jun-Hai Yong, Yisen Wang

NeurIPS 2021 · Adversarial robustness and security › Black-box and transfer attacks · approach: Query-efficient black-box attacks · 5 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
replace HSJA jump direction with tangent-point direction (TA)HSJA · ImageNet targeted, Inception-v3 @10Kmean l2 distortion: 30.425 -> 25.566Improved
replace HSJA jump direction with tangent-point direction (TA)HSJA · ImageNet untargeted, Inception-v3 @1Kmean l2 distortion: 57.295 -> 52.523Improved
replace HSJA jump direction with tangent-point direction (TA)HSJA · CIFAR-10 targeted, GDAS @10Kmean l2 distortion: 0.301 -> 0.302No clear effect
replace hemisphere with semi-ellipsoid (G-TA vs TA)TA · CIFAR-10 targeted, GDAS @10Kmean l2 distortion: 0.302 -> 0.298Improved
replace hemisphere with semi-ellipsoid (G-TA vs TA)TA · ImageNet targeted, ResNet-101 @10Kmean l2 distortion: 15.414 -> 15.633Hurt
replace hemisphere with semi-ellipsoid (G-TA vs TA)TA · CIFAR-10 untargeted, WRN-28 @10Kmean l2 distortion: 0.233 -> 0.232No clear effect
add TA to QEBA-SQEBA-S · ImageNet targeted, Inception-v3 @1Kmean l2 distortion: 79.604 -> 75.622Improved
add TA to QEBA-SQEBA-S · ImageNet targeted, Inception-v3 @10Kmean l2 distortion: 18.414 -> 17.799Improved
use TA/G-TA instead of HSJA under l_inf normHSJA · ImageNet targeted, Inception-v3 @10Kmean l_inf distortion: 0.147 -> 0.149Hurt
choose initial target-class image closest to original rather than random or farthest (ablation)Tangent Attack · ImageNet targeted, ResNet-50mean l2 distortionImproved
vary radius ratio r in G-TA (ablation)G-TA · ImageNet targeted, ResNet-50mean l2 distortionNo clear effect
jump to optimal tangent point instead of random jump direction (RandomHSJA) (ablation)HSJA · ImageNet targeted, ResNet-50mean l2 distortionImproved

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