A Smoothing Newton Method for Rank-one Matrix Recovery

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Maunu, Tyler, Abreu, Gabriel
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916872306294784
author Maunu, Tyler
Abreu, Gabriel
author_facet Maunu, Tyler
Abreu, Gabriel
contents We consider the phase retrieval problem, which involves recovering a rank-one positive semidefinite matrix from rank-one measurements. A recently proposed algorithm based on Bures-Wasserstein gradient descent (BWGD) exhibits superlinear convergence, but it is unstable, and existing theory can only prove local linear convergence for higher rank matrix recovery. We resolve this gap by revealing that BWGD implements Newton's method with a nonsmooth and nonconvex objective. We develop a smoothing framework that regularizes the objective, enabling a stable method with rigorous superlinear convergence guarantees. Experiments on synthetic data demonstrate this superior stability while maintaining fast convergence.
format Preprint
id arxiv_https___arxiv_org_abs_2507_23017
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Smoothing Newton Method for Rank-one Matrix Recovery
Maunu, Tyler
Abreu, Gabriel
Machine Learning
Optimization and Control
We consider the phase retrieval problem, which involves recovering a rank-one positive semidefinite matrix from rank-one measurements. A recently proposed algorithm based on Bures-Wasserstein gradient descent (BWGD) exhibits superlinear convergence, but it is unstable, and existing theory can only prove local linear convergence for higher rank matrix recovery. We resolve this gap by revealing that BWGD implements Newton's method with a nonsmooth and nonconvex objective. We develop a smoothing framework that regularizes the objective, enabling a stable method with rigorous superlinear convergence guarantees. Experiments on synthetic data demonstrate this superior stability while maintaining fast convergence.
title A Smoothing Newton Method for Rank-one Matrix Recovery
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2507.23017