Flat minima generalize for low-rank matrix recovery

DOI10.1093/IMAIAI/IAAE009MaRDI QIDQ6638514zbMATH OpenFDO

Authors Lijun Ding, D. Drusvyatskiy, Maryam Fazel, Zaid Harchaoui

Publication date 14 November 2024

Published in Information and Inference: A Journal of the IMA (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1093/IMAIAI/IAAE009

zbMATH Keywords

bilinear sensingcovariance matrix estimationflat minimizerslow rank matrix recoverymatrix completionmatrix sensingneural networknuclear normoverparameterizationrobust PCA

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Numerical methods for low-rank matrix approximation; matrix compression (65F55) Artificial neural networks and deep learning (68T07)

Recommendations

• Shaping the learning landscape in neural networks around wide flat minima
• Flat Minima
• Entropic gradient descent algorithms and wide flat minima*
• More is Less: Inducing Sparsity via Overparameterization
• Wide flat minima and optimal generalization in classifying high-dimensional Gaussian mixtures

Cited in

(3)

• Entropic gradient descent algorithms and wide flat minima*
• Wide flat minima and optimal generalization in classifying high-dimensional Gaussian mixtures
• More is Less: Inducing Sparsity via Overparameterization

This page was built for publication: Flat minima generalize for low-rank matrix recovery

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6638514)