AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Open-set Adversarial Defense

Rui Shao, Pramuditha Perera, Pong C. Yuen, Vishal M. Patel

ECCV 2020 · Adversarial robustness and security › Attacks on text and graphs · approach: Defenses against universal perturbations · 30 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 feature-denoising layers to the encoder (ablation)OSDN encoder · CIFAR-10 ablation (open-set)AUC-ROC: 66.1 -> 68.5Improved
add feature-denoising layers to the encoder (ablation)OSDN encoder · CIFAR-10 ablation (open-set)closed-set accuracy: 69.9 -> 70.4Improved
add decoder to the encoder (ablation)OSDN encoder · CIFAR-10 ablation (open-set)AUC-ROC: 66.1 -> 67.3Improved
add decoder to the encoder (ablation)OSDN encoder · CIFAR-10 ablation (open-set)closed-set accuracy: 69.9 -> 68.8Hurt
add denoising layers to encoder+decoder (ablation)OSDN encoder · CIFAR-10 ablation (open-set)AUC-ROC: 67.3 -> 68.2Improved
add denoising layers to encoder+decoder (ablation)OSDN encoder · CIFAR-10 ablation (open-set)closed-set accuracy: 68.8 -> 70.6Improved
add self-supervised denoising to encoder+decoder (ablation)OSDN encoder · CIFAR-10 ablation (open-set)AUC-ROC: 67.3 -> 68.5Improved
add self-supervised denoising to encoder+decoder (ablation)OSDN encoder · CIFAR-10 ablation (open-set)closed-set accuracy: 68.8 -> 71.9Improved
add self-supervised denoising on top of encoder+decoder+denoising layers (full OSDN) (ablation)OSDN encoder · CIFAR-10 ablation (open-set)AUC-ROC: 68.2 -> 69.6Improved
add self-supervised denoising on top of encoder+decoder+denoising layers (full OSDN) (ablation)OSDN encoder · CIFAR-10 ablation (open-set)closed-set accuracy: 70.6 -> 72.8Improved
use OSDN instead of feature denoisingOSDN encoder · SVHN (FGSM)AUC-ROC: 64.5 -> 71.4Improved
use OSDN instead of adversarial trainingOSDN encoder · TinyImageNet (PGD)closed-set accuracy: 40.3 -> 41.6Improved

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