AI Paper Atlas

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

ResNets Ensemble via the Feynman-Kac Formalism to Improve Natural and Robust Accuracies

Bao Wang, Zuoqiang Shi, Stanley Osher

NeurIPS 2019 · Adversarial robustness and security › Certified and empirical robustness · approach: Robust overfitting and regularization · 15 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 Gaussian noise to residual mappings and average five jointly trained ResNet20 modelsRobustly trained ResNet20 · CIFAR10natural accuracy: 75.11 -> 82.52Improved
Add Gaussian noise to residual mappings and average five jointly trained ResNet20 modelsRobustly trained ResNet20 · CIFAR10, IFGSM$^{20}$robust accuracy: 46.03 -> 51.48Improved
Increase the jointly trained noise-injected ResNet20 ensemble from one member to five (ablation)Robustly trained ResNet20 · CIFAR10natural accuracy: 77.21 -> 82.52Improved
Increase the jointly trained noise-injected ResNet20 ensemble from one member to five (ablation)Robustly trained ResNet20 · CIFAR10, IFGSM$^{20}$robust accuracy: 49.06 -> 51.48Improved
Inject noise into residual mappings and use two models instead of a single ResNet20 (ablation)Robustly trained ResNet20 · CIFAR10natural accuracy: 75.11 -> 80.34Improved
Inject noise into residual mappings and use two models instead of a single ResNet20 (ablation)Robustly trained ResNet20 · CIFAR10, IFGSM$^{20}$robust accuracy: 46.03 -> 50.06Improved
Remove residual noise from a jointly trained two-model ResNet20 ensemble (ablation)En$_2$ResNet20 · CIFAR10, IFGSM$^{20}$robust accuracy: 50.06 -> 47.28Hurt
Remove residual noise from a jointly trained two-model ResNet20 ensemble (ablation)En$_2$ResNet20 · CIFAR10natural accuracy: 80.34 -> 75.75Hurt
Integrate separately trained En$_2$ResNet110 and En$_1$WideResNet34-10En$_2$ResNet110 and En$_1$WideResNet34-10 · CIFAR10, IFGSM$^{20}$robust accuracy: 53.05 -> 57.94Improved
Replace a jointly trained two-member ResNet32 ensemble with jointly trained ResNet20 and ResNet32 (ablation)Noise-injected ResNet ensemble · CIFAR10, C&Wrobust accuracy: 68.41 -> 68.62Improved
Replace a jointly trained two-member ResNet32 ensemble with jointly trained ResNet20 and ResNet32 (ablation)Noise-injected ResNet ensemble · CIFAR100, IFGSM$^{20}$robust accuracy: 27.27 -> 27.01No clear effect
Use one noise-injected ResNet20 and one noise-injected ResNet32 rather than two noise-injected ResNet32 models (ablation)Noise-injected ResNet ensemble · CIFAR10natural accuracy: 81.46 -> 81.56No 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