Two Newton methods on the manifold of fixed-rank matrices endowed with Riemannian quotient geometries

Abstract

We consider two Riemannian geometries for the manifold $\mathcal{M}(p,m\times n)$ of all $m\times n$ matrices of rank $p$. The geometries are induced on $\mathcal{M}(p,m\times n)$ by viewing it as the base manifold of the submersion $\pi :(M,N)\mapsto M{N}^T$, selecting an adequate Riemannian metric on the total space, and turning $\pi$ into a Riemannian submersion. The theory of Riemannian submersions, an important tool in Riemannian geometry, makes it possible to obtain expressions for fundamental geometric objects on $\mathcal{M}(p,m\times n)$ and to formulate the Riemannian Newton methods on $\mathcal{M}(p,m\times n)$ induced by these two geometries. The Riemannian Newton methods admit a stronger and more streamlined convergence analysis than the Euclidean counterpart, and the computational overhead due to the Riemannian geometric machinery is shown to be mild. Potential applications include low-rank matrix completion and other low-rank matrix approximation problems.

References in corpus (3)

• Subspace Evolution and Transfer (SET) for Low-Rank Matrix Completion
• Low-rank optimization for distance matrix completion
• Low-rank optimization with trace norm penalty

Cited by in corpus (1)

• A Quadratically Convergent Algorithm for Structured Low-Rank Approximation