Improve robustness and accuracy of deep neural network with L₂, normalization

DOI10.1007/S11424-022-1326-YMaRDI QIDQ2699249zbMATH OpenOpenAlexFDO

Authors Lijia Yu, Xiao-Shan Gao

Publication date 26 April 2023

Published in Journal of Systems Science and Complexity (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2010.04912

zbMATH Keywords

\(L_{2,\infty}\) normalizationdeep neural networkglobal robustness measureover-fittingRademacher complexitysmooth DNN

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07)

Recommendations

• Sparse deep neural networks using \(L_{1,\infty}\)-weight normalization
• Improved spectral norm regularization for neural networks
• Theoretical investigation of generalization bounds for adversarial learning of deep neural networks
• Probabilistic robustness estimates for feed-forward neural networks
• Adversarial defense via the data-dependent activation, total variation minimization, and adversarial training

Cites work

• Deep learning
• scientific article; zbMATH DE number 6378127 (Why is no real title available?)
• scientific article; zbMATH DE number 3252945 (Why is no real title available?)
• Improve robustness of sparse PCA by \(L_{1}\)-norm maximization
• Sparse deep neural networks using \(L_{1,\infty}\)-weight normalization

Cited in

(4)

• A new initialization method based on normed statistical spaces in deep networks
• Preface to the special topic on computer mathematics
• Improved spectral norm regularization for neural networks
• Enhancing multivariate time series forecasting: a novel approach with Mallows model averaging and graph neural networks

This page was built for publication: Improve robustness and accuracy of deep neural network with \(L_{2,\infty}\) normalization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2699249)