Entropic regularization of Wasserstein distance between infinite-dimensional Gaussian measures and Gaussian processes

Abstract

This work studies the entropic regularization formulation of the 2-Wasserstein distance on an infinite-dimensional Hilbert space, in particular for the Gaussian setting. We first present the Minimum Mutual Information property, namely the joint measures of two Gaussian measures on Hilbert space with the smallest mutual information are joint Gaussian measures. This is the infinite-dimensional generalization of the Maximum Entropy property of Gaussian densities on Euclidean space. We then give closed form formulas for the optimal entropic transport plan, entropic 2-Wasserstein distance, and Sinkhorn divergence between two Gaussian measures on a Hilbert space, along with the fixed point equations for the barycenter of a set of Gaussian measures. Our formulations fully exploit the regularization aspect of the entropic formulation and are valid both in singular and nonsingular settings. In the infinite-dimensional setting, both the entropic 2-Wasserstein distance and Sinkhorn divergence are Fréchet differentiable, in contrast to the exact 2-Wasserstein distance, which is not differentiable. Our Sinkhorn barycenter equation is new and always has a unique solution. In contrast, the finite-dimensional barycenter equation for the entropic 2-Wasserstein distance fails to generalize to the Hilbert space setting. In the setting of reproducing kernel Hilbert spaces (RKHS), our distance formulas are given explicitly in terms of the corresponding kernel Gram matrices, providing an interpolation between the kernel Maximum Mean Discrepancy (MMD) and the kernel 2-Wasserstein distance.

arXiv commentrevised version, 90 pages, to be published in Journal of Theoretical Probability

References in corpus (9)

• Non-Euclidean statistics for covariance matrices, with applications to diffusion tensor imaging
• Stochastic Optimization for Large-scale Optimal Transport
• Log-Determinant Divergences Revisited: Alpha--Beta and Gamma Log-Det Divergences
• Entropic Optimal Transport between Unbalanced Gaussian Measures has a Closed Form
• Sinkhorn AutoEncoders
• Benamou-Brenier and duality formulas for the entropic cost on $RC{D}^{\ast }(K,N)$ spaces
• Debiased Sinkhorn barycenters
• The statistical effect of entropic regularization in optimal transportation
• Alpha Procrustes metrics between positive definite operators: a unifying formulation for the Bures-Wasserstein and Log-Euclidean/Log-Hilbert-Schmidt metrics

Cited by in corpus (4)

• Unbalanced Multi-Marginal Optimal Transport
• Convergence and finite sample approximations of entropic regularized Wasserstein distances in Gaussian and RKHS settings
• Estimating 2-Sinkhorn Divergence between Gaussian Processes from Finite-Dimensional Marginals
• Estimation of Riemannian distances between covariance operators and Gaussian processes