Asymptotic Escape of Spurious Critical Points on the Low-rank Matrix Manifold

Abstract

We show that on the manifold of fixed-rank and symmetric positive semi-definite matrices, the Riemannian gradient descent algorithm almost surely escapes some spurious critical points on the boundary of the manifold. Our result is the first to partially overcome the incompleteness of the low-rank matrix manifold without changing the vanilla Riemannian gradient descent algorithm. The spurious critical points are some rank-deficient matrices that capture only part of the eigen components of the ground truth. Unlike classical strict saddle points, they exhibit very singular behavior. We show that using the dynamical low-rank approximation and a rescaled gradient flow, some of the spurious critical points can be converted to classical strict saddle points in the parameterized domain, which leads to the desired result. Numerical experiments are provided to support our theoretical findings.

References in corpus (7)

• Nonconvex Optimization Meets Low-Rank Matrix Factorization: An Overview
• Gradient Descent Converges to Minimizers
• First-order Methods Almost Always Avoid Saddle Points
• Efficiently escaping saddle points on manifolds
• Escaping from saddle points on Riemannian manifolds
• Fast Global Convergence for Low-rank Matrix Recovery via Riemannian Gradient Descent with Random Initialization
• Analysis of Asymptotic Escape of Strict Saddle Sets in Manifold Optimization

Cited by in corpus (1)

• Finding stationary points on bounded-rank matrices: A geometric hurdle and a smooth remedy