Wasserstein Dictionary Learning: Optimal Transport-based unsupervised non-linear dictionary learning

Abstract

This paper introduces a new nonlinear dictionary learning method for histograms in the probability simplex. The method leverages optimal transport theory, in the sense that our aim is to reconstruct histograms using so-called displacement interpolations (a.k.a. Wasserstein barycenters) between dictionary atoms; such atoms are themselves synthetic histograms in the probability simplex. Our method simultaneously estimates such atoms, and, for each datapoint, the vector of weights that can optimally reconstruct it as an optimal transport barycenter of such atoms. Our method is computationally tractable thanks to the addition of an entropic regularization to the usual optimal transportation problem, leading to an approximation scheme that is efficient, parallel and simple to differentiate. Both atoms and weights are learned using a gradient-based descent method. Gradients are obtained by automatic differentiation of the generalized Sinkhorn iterations that yield barycenters with entropic smoothing. Because of its formulation relying on Wasserstein barycenters instead of the usual matrix product between dictionary and codes, our method allows for nonlinear relationships between atoms and the reconstruction of input data. We illustrate its application in several different image processing settings.

Published in SIAM SIIMS. 46 pages, 24 figures

References in corpus (1)

• Super-resolution method using sparse regularization for point-spread function recovery

Cited by in corpus (12)

• Point spread function modelling for astronomical telescopes: a review focused on weak gravitational lensing studies
• 21st Century Statistical and Computational Challenges in Astrophysics
• Euclid: Nonparametric point spread function field recovery through interpolation on a graph Laplacian
• Multi-CCD Point Spread Function Modelling
• Rethinking data-driven point spread function modeling with a differentiable optical model
• A general drift estimation procedure for stochastic differential equations with additive fractional noise
• Ground Metric Learning on Graphs
• Multi-Source Domain Adaptation through Dataset Dictionary Learning in Wasserstein Space
• Wasserstein Index Generation Model: Automatic Generation of Time-series Index with Application to Economic Policy Uncertainty
• On the metric property of quantum Wasserstein divergences
• Isometries of the qubit state space with respect to quantum Wasserstein distances
• A Family of Pairwise Multi-Marginal Optimal Transports that Define a Generalized Metric