Infinitesimal behavior of quadratically regularized optimal transport and its relation with the porous medium equation

A Garriz-Molina, A González-Sanz… - arxiv preprint arxiv …, 2024 - arxiv.org
The quadratically regularized optimal transport problem has recently been considered in
various applications where the coupling needs to be\emph {sparse}, ie, the density of the …

Sparsity of quadratically regularized optimal transport: Scalar case

A González-Sanz, M Nutz - arxiv preprint arxiv:2410.03353, 2024 - arxiv.org
The quadratically regularized optimal transport problem is empirically known to have sparse
solutions: its optimal coupling $\pi_ {\varepsilon} $ has sparse support for small …

Quantitative convergence of quadratically regularized linear programs

A González-Sanz, M Nutz - arxiv preprint arxiv:2408.04088, 2024 - arxiv.org
Linear programs with quadratic regularization are attracting renewed interest due to their
applications in optimal transport: unlike entropic regularization, the squared-norm penalty …

Nonlinear inverse optimal transport: Identifiability of the transport cost from its marginals and optimal values

A González-Sanz, M Groppe, A Munk - SIAM Journal on Mathematical …, 2024 - SIAM
The inverse optimal transport problem is to find the underlying cost function from the
knowledge of optimal transport plans. While this amounts to solving a linear inverse …

Monotonicity in quadratically regularized linear programs

A González-Sanz, M Nutz, AR Valdevenito - arxiv preprint arxiv …, 2024 - arxiv.org
In optimal transport, quadratic regularization is a sparse alternative to entropic
regularization: the solution measure tends to have small support. Computational experience …

Sparsity of quadratically regularized optimal transport: Bounds on concentration and bias

J Wiesel, X Xu - arxiv preprint arxiv:2410.03425, 2024 - arxiv.org
We study the quadratically regularized optimal transport (QOT) problem for quadratic cost
and compactly supported marginals $\mu $ and $\nu $. It has been empirically observed that …

Entropic Optimal Transport Eigenmaps for Nonlinear Alignment and Joint Embedding of High-Dimensional Datasets

B Landa, Y Kluger, R Ma - arxiv preprint arxiv:2407.01718, 2024 - arxiv.org
Embedding high-dimensional data into a low-dimensional space is an indispensable
component of data analysis. In numerous applications, it is necessary to align and jointly …

The entropic optimal (self-) transport problem: Limit distributions for decreasing regularization with application to score function estimation

G Mordant - arxiv preprint arxiv:2412.12007, 2024 - arxiv.org
Westudythestatisticalpropertiesoftheentropi… (self) transport problem for smooth probability
measures. We provide an accurate description of the limit distribution for entropic (self-) …

Regularised optimal self-transport is approximate Gaussian mixture maximum likelihood

G Mordant - International Conference on Soft Methods in Probability …, 2024 - Springer
We investigate the link between regularised self-transport problems and maximum
likelihood estimation in Gaussian mixture models (GMM). This link suggests that self …

Quadratically Regularized Optimal Transport: Existence and Multiplicity of Potentials

M Nutz - arxiv preprint arxiv:2404.06847, 2024 - arxiv.org
The optimal transport problem with quadratic regularization is useful when sparse couplings
are desired. The density of the optimal coupling is described by two functions called …