On Riemannian Optimization over Positive Definite Matrices with the Bures-Wasserstein Geometry

Abstract

In this paper, we comparatively analyze the Bures-Wasserstein (BW) geometry with the popular Affine-Invariant (AI) geometry for Riemannian optimization on the symmetric positive definite (SPD) matrix manifold. Our study begins with an observation that the BW metric has a linear dependence on SPD matrices in contrast to the quadratic dependence of the AI metric. We build on this to show that the BW metric is a more suitable and robust choice for several Riemannian optimization problems over ill-conditioned SPD matrices. We show that the BW geometry has a non-negative curvature, which further improves convergence rates of algorithms over the non-positively curved AI geometry. Finally, we verify that several popular cost functions, which are known to be geodesic convex under the AI geometry, are also geodesic convex under the BW geometry. Extensive experiments on various applications support our findings.

References in corpus (7)

• Non-Euclidean statistics for covariance matrices, with applications to diffusion tensor imaging
• Sparse Coding and Dictionary Learning for Symmetric Positive Definite Matrices: A Kernel Approach
• A Riemannian Network for SPD Matrix Learning
• Regression on fixed-rank positive semidefinite matrices: a Riemannian approach
• On Wasserstein geometry of the space of Gaussian measures
• Is affine invariance well defined on SPD matrices? A principled continuum of metrics
• Variance reduction for Riemannian non-convex optimization with batch size adaptation