EnResNet: ResNets Ensemble via the Feynman--Kac Formalism for Adversarial Defense and Beyond

DOI10.1137/19M1265302MaRDI QIDQ5037566zbMATH OpenFDO

Authors Bao Wang, Binjie Yuan, Zuoqiang Shi, Stanley Osher

Publication date 1 March 2022

Published in SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1811.10745

zbMATH Keywords

adversarial defenseconvection-diffusion equationdeep learningFeynman-Kac formula

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07)

Recommendations

• Robustifying models against adversarial attacks by Langevin dynamics
• Generating universal adversarial perturbation with ResNet
• VectorDefense: Vectorization as a Defense to Adversarial Examples

Cites work

• A proposal on machine learning via dynamical systems
• Adversarial defense via the data-dependent activation, total variation minimization, and adversarial training
• Graph interpolating activation improves both natural and robust accuracies in data-efficient deep learning
• scientific article; zbMATH DE number 54145 (Why is no real title available?)
• scientific article; zbMATH DE number 3277871 (Why is no real title available?)
• On Distributions of Certain Wiener Functionals
• Stable architectures for deep neural networks
• The relative performance of ensemble methods with deep convolutional neural networks for image classification

Cited in

(9)

• Adversarial defense via the data-dependent activation, total variation minimization, and adversarial training
• Generating universal adversarial perturbation with ResNet
• Wavelet regularization benefits adversarial training
• A fast saddle-point dynamical system approach to robust deep learning
• Accuracy and architecture studies of residual neural network method for ordinary differential equations
• Deep learning via dynamical systems: an approximation perspective
• ResNets Ensemble via the Feynman-Kac Formalism to Improve Natural and Robust Accuracies
• Robustness Verification for Classifier Ensembles
• Deep neural network for solving differential equations motivated by Legendre-Galerkin approximation

Describes a project that uses

Uses Software

• ImageNet
• AlexNet
• CIFAR
• torchdiffeq
• Adam
• GitHub

This page was built for publication: EnResNet: ResNets Ensemble via the Feynman--Kac Formalism for Adversarial Defense and Beyond

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5037566)