The non-convex geometry of low-rank matrix optimization

DOI10.1093/imaiai/iay003zbMath1476.90245arXiv1611.03060OpenAlexW2789612570MaRDI QIDQ5242864

Zhihui Zhu, Qiuwei Li, Gongguo Tang

Publication date: 7 November 2019

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

Full work available at URL: https://arxiv.org/abs/1611.03060

zbMATH Keywords

global convergencenegative curvaturematrix factorizationnuclear normlow rankweighted PCABurer-Monteiro1-bit matrix recoverystrict saddle property

Mathematics Subject Classification ID

Convex programming (90C25) Convexity of real functions in one variable, generalizations (26A51)

Related Items (12)

Model-free Nonconvex Matrix Completion: Local Minima Analysis and Applications in Memory-efficient Kernel PCA ⋮ Nonconvex Low-Rank Tensor Completion from Noisy Data ⋮ Column $\ell_{2,0}$-Norm Regularized Factorization Model of Low-Rank Matrix Recovery and Its Computation ⋮ Spurious Valleys, NP-Hardness, and Tractability of Sparse Matrix Factorization with Fixed Support ⋮ Low-Rank Univariate Sum of Squares Has No Spurious Local Minima ⋮ A matrix nonconvex relaxation approach to unconstrained binary polynomial programs ⋮ The global optimization geometry of shallow linear neural networks ⋮ Nonconvex Robust Low-Rank Matrix Recovery ⋮ Error bound of critical points and KL property of exponent 1/2 for squared F-norm regularized factorization ⋮ Rank Optimality for the Burer--Monteiro Factorization ⋮ Provable accelerated gradient method for nonconvex low rank optimization ⋮ On the Landscape of Synchronization Networks: A Perspective from Nonconvex Optimization

This page was built for publication: The non-convex geometry of low-rank matrix optimization