Wasserstein Riemannian Geometry of Positive Definite Matrices
arXiv:1801.09269
Abstract
The Wasserstein distance on multivariate non-degenerate Gaussian densities is a Riemannian distance. After reviewing the properties of the distance and the metric geodesic, we present an explicit form of the Riemannian metrics on positive-definite matrices and compute its tensor form with respect to the trace inner product. The tensor is a matrix which is the solution to a Lyapunov equation. We compute the explicit formula for the Riemannian exponential, the normal coordinates charts and the Riemannian gradient. Finally, the Levi-Civita covariant derivative is computed in matrix form together with the differential equation for the parallel transport. While all computations are given in matrix form, nonetheless we discuss also the use of a special moving frame.
arXiv commentRevised version. Introduction extended. References added
References in corpus (2)
Cited by in corpus (8)
- Generalizing Point Embeddings using the Wasserstein Space of Elliptical Distributions
- Entropic Optimal Transport between Unbalanced Gaussian Measures has a Closed Form
- Information-geometry of physics-informed statistical manifolds and its use in data assimilation
- Optimal transport natural gradient for statistical manifolds with continuous sample space
- Information Newton's flow: second-order optimization method in probability space
- Kernelized Wasserstein Natural Gradient
- Dynamical Wasserstein Barycenters for Time-series Modeling
- Hessian transport gradient flows