Random Matrix-Improved Estimation of the Wasserstein Distance between two Centered Gaussian Distributions

Code

• github.com/maliktiomoko/RMTWasserstein
  OfficialIn papernone★ 0

Abstract

This article proposes a method to consistently estimate functionals 1p_i=1^pf(_i(C_1C_2)) of the eigenvalues of the product of two covariance matrices C_1,C_2R^p p based on the empirical estimates _i( C_1 C_2) ( C_a=1n_a_i=1^n_a x_i^(a)x_i^(a) T), when the size p and number n_a of the (zero mean) samples x_i^(a) are similar. As a corollary, a consistent estimate of the Wasserstein distance (related to the case f(t)=t) between centered Gaussian distributions is derived. The new estimate is shown to largely outperform the classical sample covariance-based `plug-in' estimator. Based on this finding, a practical application to covariance estimation is then devised which demonstrates potentially significant performance gains with respect to state-of-the-art alternatives.

Reproductions