AI Paper Atlas

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

A Simple Way to Make Neural Networks Robust Against Diverse Image Corruptions

Evgenia Rusak, Lukas Schott, Roland S. Zimmermann, Julian Bitterwolf, Oliver Bringmann, Matthias Bethge, Wieland Brendel

ECCV 2020 · Adversarial robustness and security › Attacks on text and graphs · approach: Biologically inspired robust vision

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 experience replay of earlier noise generators (ablation)ANT 1x1 (ResNet50) · ImageNet-C (all, Top-1)accuracy (%): 48.9 -> 51.1Improved
add experience replay of earlier noise generators (ablation)ANT 1x1 (ResNet50) · ImageNet-C w/o Noise categories (Top-1)accuracy (%): 46.5 -> 47.7Improved
use a single tuned sigma 0.5 instead of sampling sigma from a set (ablation)Gaussian noise training (ResNet50) · ImageNet-C w/o Noise categories (Top-1)accuracy (%): 45.2 -> 47.1Improved
use adversarial noise training instead of Gaussian noise trainingResNet50 · ImageNet-C (all, Top-1)accuracy (%): 49.4 -> 51.1Improved
combine ANT 1x1 with stylized data (SIN)ANT 1x1 (ResNet50) · ImageNet-C (all, Top-1)accuracy (%): 51.1 -> 52.2Improved
combine ANT 3x3 with stylized data (SIN)ANT 3x3 (ResNet50) · ImageNet-C (all, Top-1)accuracy (%): 50.4 -> 52.6Improved
use locally correlated 3x3 noise generator instead of 1x1ANT (ResNet50) · ImageNet-C (all, Top-1)accuracy (%): 51.1 -> 50.4Mixed
train against adversarial noise sampled randomly instead of the generator-driven scheme (ablation)ANT 1x1 (ResNet50) · ImageNet-C w/o Noise categories (Top-1)accuracy (%): 47.7 -> 47.3Hurt
apply regular adversarial trainingResNet50 · ImageNet-C (all, Top-1)accuracy (%): 39.2 -> 29.1Hurt
apply regular adversarial trainingResNet152 · ImageNet-C (all, Top-1)accuracy (%): 45.0 -> 35.0Hurt
apply universal adversarial trainingAlexNet · ImageNet-C (all, Top-1)accuracy (%): 21.1 -> 22.2No clear effect
fine-tune with ANT 1x1DenseNet121 · ImageNet-C (all, Top-1)accuracy (%): 42.1 -> 50.3Improved

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