AI Paper Atlas

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

Once a MAN: Towards Multi-Target Attack via Learning Multi-Target Adversarial Network Once

Jiangfan Han, Xiaoyi Dong, Ruimao Zhang, Dongdong Chen, Weiming Zhang, Nenghai Yu, Ping Luo, Xiaogang Wang

ICCV 2019 · Adversarial robustness and security › Black-box and transfer attacks · approach: Transfer attacks on face recognition · 33 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
train with random target labels (multi-target) instead of one fixed target (ablation)MANr · CIFAR10 Res32 to VGG16 transferattack success rate (%): 58.71 -> 67.69Improved
jointly train against VGG16 and ResNet32 (model ensemble) (ablation)MANr · CIFAR10 Res32 to VGG19 transferattack success rate (%): 72.26 -> 98.07Improved
jointly train against VGG16 and ResNet32 (model ensemble) (ablation)MANc · CIFAR10 Res32 to Res14 transferattack success rate (%): 81.01 -> 96.79Improved
use the channel-recalibration label fusion (MANr) instead of concatenation (MANc) (ablation)MAN · CIFAR10 multi-targetattack success rate (%): 99.14 -> 99.50Improved
use concatenation label fusion (MANc) instead of recalibration (MANr) (ablation)MAN · ImageNet multi-target, Res152 to Res101attack success rate (%): 49.69 -> 55.39Improved
condition generator on target label (MANc) versus single-target GAPMANc · ImageNet VGG16 to VGG19 transferattack success rate (%): 70.83 -> 85.79Improved
fine-tune with MANc adversarial samples instead of GAP10 samplesMANc · CIFAR10 MI-FGSM strength 16sqrt(N)classification accuracy (%): 79.08 -> 81.14Improved
fine-tune with adversarial samples from ten GAP models instead of one (ablation)GAP · CIFAR10 MI-FGSM strength 16sqrt(N)classification accuracy (%): 63.40 -> 79.08Improved
fine-tune ResNet32 with MANc adversarial samplesResNet32 · CIFAR10 MI-FGSM strength 16sqrt(N)classification accuracy (%): 10.58 -> 81.14Improved
increase perturbation threshold epsilon from very small values (ablation)MAN · CIFAR10 VGG16attack success rateImproved
increase perturbation threshold epsilon beyond 12sqrt(N) (ablation)MAN · CIFAR10 VGG16attack success rateHurt
use a larger reconstruction weight alpha (ablation)MAN · CIFAR10 VGG16attack success rateMixed

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