On Geometric Connections of Embedded and Quotient Geometries in Riemannian Fixed-rank Matrix Optimization

Abstract

In this paper, we propose a general procedure for establishing the landscape connections of a Riemannian optimization problem under the embedded and quotient geometries. By applying the general procedure to the fixed-rank positive semidefinite (PSD) and general matrix optimization, we establish an exact Riemannian gradient connection under two geometries at every point on the manifold and sandwich inequalities between the spectra of Riemannian Hessians at Riemannian first-order stationary points (FOSPs). These results immediately imply an equivalence on the sets of Riemannian FOSPs, Riemannian second-order stationary points (SOSPs) and strict saddles of fixed-rank matrix optimization under the embedded and the quotient geometries. To the best of our knowledge, this is the first geometric landscape connection between the embedded and the quotient geometries for fixed-rank matrix optimization and it provides a concrete example on how these two geometries are connected in Riemannian optimization. In addition, the effects of the Riemannian metric and quotient structure on the landscape connection are discussed. We also observe an algorithmic connection for fixed-rank matrix optimization under two geometries with some specific Riemannian metrics. A number of novel ideas and technical ingredients including a unified treatment for different Riemannian metrics and new horizontal space representations under quotient geometries are developed to obtain our results. The results in this paper deepen our understanding of geometric connections of Riemannian optimization under different Riemannian geometries and provide a few new theoretical insights to unanswered questions in the literature.

On the analysis of optimization with fixed-rank matrices: a quotient geometric view

“…Proposition 3.1 is one of the results in a chapter of the thesis [12, §5.2.3]. In a more recent work [24,Remark 9], the same result is given as an example of the geometric connections of embedded and quotient geometries in Riemannian fixed-rank matrix optimization.…”

“…We prove the spectral lower bound of Hessf (M ⋆ ) as follows. First, the exponential map at Y := M ⋆ has the following expression (e.g., [3], [38, Proposition A1], [37, Appendix A]), (4.24) Exp 24), it follows that the quadratic function f (4.1) satisfies, for any…”