Riemannian Langevin Monte Carlo schemes for sampling PSD matrices with fixed rank

Abstract
This paper introduces two explicit schemes to sample matrices from Gibbs distributions on , the manifold of real positive semi-definite (PSD) matrices of size and rank . Given an energy function and certain Riemannian metrics on , these schemes rely on an Euler-Maruyama discretization of the Riemannian Langevin equation (RLE) with Brownian motion on the manifold. We present numerical schemes for RLE under two fundamental metrics on : (a) the metric obtained from the embedding of ; and (b) the Bures-Wasserstein metric corresponding to quotient geometry. We also provide examples of energy functions with explicit Gibbs distributions that allow numerical validation of these schemes.
View on arXivComments on this paper