AI Paper Atlas

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

Adversarial Learning With Margin-Based Triplet Embedding Regularization

Yaoyao Zhong, Weihong Deng

ICCV 2019 · Adversarial robustness and security › Certified and empirical robustness · approach: Robust overfitting and regularization · 48 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
add MTER regularizationSoftmax ResNet-18 · MNIST BIMaccuracy under attack (%): 0.0 -> 98.7Improved
add MTER regularizationArcFace ResNet-18 · MNIST ITGSMaccuracy under attack (%): 24.7 -> 95.3Improved
replace adversarial training with MTERSoftmax ResNet-18 · MNIST BIMaccuracy under attack (%): 3.1 -> 98.7Improved
replace adversarial training with MTERSoftmax ResNet-18 · MNIST FGSMaccuracy under attack (%): 99.9 -> 96.8Hurt
add MTER regularizationSoftmax ResNet-18 · MNIST cleanaccuracy (%): 99.6 -> 99.5No clear effect
add MTER finetuningArcFace ResNet-50 · Face disguise IFTGSM eps=10 (CASIA-WebFace)hit rate (%): 100.0 -> 27.4Improved
add MTER finetuningArcFace ResNet-50 · Face disguise IFTGSM eps=10 (MS1M)hit rate (%): 100.0 -> 61.4Improved
finetune on MS1M+VGGFace2 instead of MS1M only with MTERArcFace ResNet-50 · Face disguise IFTGSM eps=10hit rate (%): 61.4 -> 19.5Improved
add MTER finetuningArcFace ResNet-50 · LFW (trained on CASIA-WebFace)accuracy (%): 99.5 -> 99.5No clear effect
replace adversarial finetuning with MTERArcFace ResNet-50 · Face disguise IFTGSM eps=10 (CASIA-WebFace)hit rate (%): 49.5 -> 27.4Improved
replace adversarial finetuning with MTERArcFace ResNet-50 · Face disguise IFTGSM eps=10 (MS1M)hit rate (%): 62.6 -> 61.4No clear effect

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