Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published December 22, 2022 | Submitted
Report Open

Fast Global Convergence for Low-rank Matrix Recovery via Riemannian Gradient Descent with Random Initialization

Abstract

In this paper, we propose a new global analysis framework for a class of low-rank matrix recovery problems on the Riemannian manifold. We analyze the global behavior for the Riemannian optimization with random initialization. We use the Riemannian gradient descent algorithm to minimize a least squares loss function, and study the asymptotic behavior as well as the exact convergence rate. We reveal a previously unknown geometric property of the low-rank matrix manifold, which is the existence of spurious critical points for the simple least squares function on the manifold. We show that under some assumptions, the Riemannian gradient descent starting from a random initialization with high probability avoids these spurious critical points and only converges to the ground truth in nearly linear convergence rate, i.e. O(log(1/ϵ) + log(n)) iterations to reach an ϵ-accurate solution. We use two applications as examples for our global analysis. The first one is a rank-1 matrix recovery problem. The second one is a generalization of the Gaussian phase retrieval problem. It only satisfies the weak isometry property, but has behavior similar to that of the first one except for an extra saddle set. Our convergence guarantee is nearly optimal and almost dimension-free, which fully explains the numerical observations. The global analysis can be potentially extended to other data problems with random measurement structures and empirical least squares loss functions.

Additional Information

Attribution 4.0 International (CC BY 4.0) This research was in part supported by NSF Grants DMS P2259068 and DMS P2259075. We would also like to thank Prof. Jian-feng Cai for helpful suggestions.

Attached Files

Submitted - 2012.15467.pdf

Files

2012.15467.pdf
Files (924.1 kB)
Name Size Download all
md5:8155a3f6d856ebda7f23f8b98b1ba4cc
924.1 kB Preview Download

Additional details

Created:
August 20, 2023
Modified:
October 24, 2023