Sobolev regularity for optimal transport maps of non-convex planar domains

Connor Mooney,Arghya Rakshit

2023-06-14

Abstract:We prove a sharp global $W^{2,\,p}$ estimate for potentials of optimal transport maps that take a certain class of non-convex planar domains to convex ones.

Analysis of PDEs

What problem does this paper attempt to address?

Pointwise Schauder estimates for optimal transport maps of rough densities

Arghya Rakshit

2024-10-09

Abstract:We prove a pointwise $C^{2,\,\alpha}$ estimate for the potential of the optimal transport map in the case that the densities are only close to a constant in a certain $L^p$ sense. To do so we use the variational approach recently developed by Goldman-Otto[ Ann. Sci. Éc. Norm. Supér.(2020)].

Analysis of PDEs
Partial $W^{2,p}$ regularity for optimal transport maps

Shibing Chen,Alessio Figalli

DOI: https://doi.org/10.48550/arXiv.1606.05173

2016-06-16

Analysis of PDEs

Abstract:We prove that, in the optimal transportation problem with general costs and positive continuous densities, the potential function is always of class $W^{2,p}_{loc}$ for any $p \geq 1$ outside of a closed singular set of measure zero. We also establish global $W^{2,p}$ estimates when the cost is a small perturbation of the quadratic cost. The latter result is new even when the cost is exactly the quadratic cost.
Global Regularity of Optimal Mappings in Non-Convex Domains

Shibing Chen,Jiakun Liu,Xu-Jia Wang

DOI: https://doi.org/10.1007/s11425-018-9465-8

2019-01-01

Science China Mathematics

Abstract:In this paper, we establish a global regularity result for the optimal transport problem with the quadratic cost, where the domains may not be convex. This result is obtained by a perturbation argument, using a recent global regularity of optimal transportation in convex domains by the authors.
Central Limit Theorems for Smooth Optimal Transport Maps

Tudor Manole,Sivaraman Balakrishnan,Jonathan Niles-Weed,Larry Wasserman

2024-09-17

Abstract:One of the central objects in the theory of optimal transport is the Brenier map: the unique monotone transformation which pushes forward an absolutely continuous probability law onto any other given law. A line of recent work has analyzed $L^2$ convergence rates of plugin estimators of Brenier maps, which are defined as the Brenier map between density estimators of the underlying distributions. In this work, we show that such estimators satisfy a pointwise central limit theorem when the underlying laws are supported on the flat torus of dimension $d \geq 3$. We also derive a negative result, showing that these estimators do not converge weakly in $L^2$ when the dimension is sufficiently large. Our proofs hinge upon a quantitative linearization of the Monge-Ampère equation, which may be of independent interest. This result allows us to reduce our problem to that of deriving limit laws for the solution of a uniformly elliptic partial differential equation with a stochastic right-hand side, subject to periodic boundary conditions.

Probability,Analysis of PDEs,Statistics Theory
On strict convexity and $C^1$ regularity of potential functions in optimal transportation

Neil S. Trudnger,Xu-Jia Wang

DOI: https://doi.org/10.48550/arXiv.math/0702807

2007-02-27

Abstract:This note concerns the relationship between conditions on cost functions and domains and the convexity properties of potentials in optimal transportation and the continuity of the associated optimal mappings. In particular, we prove that if the cost function satisfies the condition (A3), introduced in our previous work with Xinan Ma, the densities and their reciprocals are bounded and the target domain is convex with respect to the cost function, then the potential is continuously differentiable and its dual potential strictly concave with respect to the cost function. Our result extends, by different and more direct proof, similar results of Loeper proved by approximation from our earlier work on global regularity.

Analysis of PDEs
On optimal transport of matrix-valued measures

Yann Brenier,Dmitry Vorotnikov

DOI: https://doi.org/10.48550/arXiv.1808.05064

2019-07-15

Abstract:We suggest a new way of defining optimal transport of positive-semidefinite matrix-valued measures. It is inspired by a recent rendering of the incompressible Euler equations and related conservative systems as concave maximization problems. The main object of our attention is the Kantorovich-Bures metric space, which is a matricial analogue of the Wasserstein and Hellinger-Kantorovich metric spaces. We establish some topological, metric and geometric properties of this space, which includes the existence of the optimal transportation path.

Optimization and Control,Functional Analysis,Metric Geometry
Sharp Sobolev inequalities on noncompact Riemannian manifolds with via optimal transport theory

Alexandru Kristály

DOI: https://doi.org/10.1007/s00526-024-02810-9

2024-08-19

Calculus of Variations and Partial Differential Equations

Abstract:In their seminal work, Cordero-Erausquin, Nazaret and Villani (Adv Math 182(2):307-332, 2004) proved sharp Sobolev inequalities in Euclidean spaces via Optimal Transport , raising the question whether their approach is powerful enough to produce sharp Sobolev inequalities also on Riemannian manifolds. By using -optimal transport approach, the compact case has been successfully treated by Cavalletti and Mondino (Geom Topol 21:603-645, 2017), even on metric measure spaces verifying the synthetic lower Ricci curvature bound. In the present paper we affirmatively answer the above question for noncompact Riemannian manifolds with non-negative Ricci curvature; namely, by using Optimal Transport theory with quadratic distance cost, sharp -Sobolev and -logarithmic Sobolev inequalities (both for and ) are established, where the sharp constants contain the asymptotic volume ratio arising from precise asymptotic properties of the Talentian and Gaussian bubbles, respectively. As a byproduct, we give an alternative, elementary proof to the main result of do Carmo and Xia (Math 140:818-826, 2004) and subsequent results, concerning the quantitative volume non-collapsing estimates on Riemannian manifolds with non-negative Ricci curvature that support Sobolev inequalities.

mathematics, applied
Sharp Sobolev inequalities on noncompact Riemannian manifolds with ${\sf Ric}\geq 0$ via Optimal Transport theory

Alexandru Kristály

2024-07-26

Abstract:In their seminal work, Cordero-Erausquin, Nazaret and Villani [Adv. Math., 2004] proved sharp Sobolev inequalities in Euclidean spaces via Optimal Mass Transportation, raising the question whether their approach is powerful enough to produce sharp Sobolev inequalities also on Riemannian manifolds. In this paper we affirmatively answer their question for Riemannian manifolds with non-negative Ricci curvature; namely, by using Optimal Mass Transportation with quadratic distance cost, sharp $L^p$-Sobolev and $L^p$-logarithmic Sobolev inequalities (both for $p>1$ and $p=1$) are established, where the optimal constants contain the asymptotic volume growth arising from precise asymptotic properties of the Talentian and Gaussian bubbles. As a byproduct, we give an alternative, elementary proof to the main result of do Carmo and Xia [Compos. Math., 2004] (and subsequent results) concerning the quantitative volume non-collapsing estimates on Riemannian manifolds with non-negative Ricci curvature that support certain Sobolev inequalities.

Analysis of PDEs,Differential Geometry,Functional Analysis
On Optimal Transport Maps Between 1 /d-Concave Densities

Guillaume Carlier,Alessio Figalli,Filippo Santambrogio

2024-04-08

Abstract:In this paper, we extend the scope of Caffarelli's contraction theorem, which provides a measure of the Lipschitz constant for optimal transport maps between log-concave probability densities in $\R^d$. Our focus is on a broader category of densities, specifically those that are $\nicefrac{1}{d}$-concave and can be represented as $V^{-d}$, where $V$ is convex. By setting appropriate conditions, we derive linear or sublinear limitations for the optimal transport map. This leads us to a comprehensive Lipschitz estimate that aligns with the principles established in Caffarelli's theorem.

Analysis of PDEs
Weak limits of entropy regularized Optimal Transport; potentials, plans and divergences

Alberto Gonzalez-Sanz,Jean-Michel Loubes,Jonathan Niles-Weed

2024-06-05

Abstract:This work deals with the asymptotic distribution of both potentials and couplings of entropic regularized optimal transport for compactly supported probabilities in $\R^d$. We first provide the central limit theorem of the Sinkhorn potentials -- the solutions of the dual problem -- as a Gaussian process in $\Cs$. Then we obtain the weak limits of the couplings -- the solutions of the primal problem -- evaluated on integrable functions, proving a conjecture of \cite{ChaosDecom}. In both cases, their limit is a real Gaussian random variable. Finally we consider the weak limit of the entropic Sinkhorn divergence under both assumptions $H_0:\ {\rm P}={\rm Q}$ or $H_1:\ {\rm P}\neq{\rm Q}$. Under $H_0$ the limit is a quadratic form applied to a Gaussian process in a Sobolev space, while under $H_1$, the limit is Gaussian. We provide also a different characterisation of the limit under $H_0$ in terms of an infinite sum of an i.i.d. sequence of standard Gaussian random variables. Such results enable statistical inference based on entropic regularized optimal transport.

Probability,Statistics Theory
Quadratically Regularized Optimal Transport: nearly optimal potentials and convergence of discrete Laplace operators

Gilles Mordant,Stephen Zhang

DOI: https://doi.org/10.48550/arXiv.2212.00103

2022-11-21

Abstract:We consider the conjecture proposed in Matsumoto, Zhang and Schiebinger (2022) suggesting that optimal transport with quadratic regularisation can be used to construct a graph whose discrete Laplace operator converges to the Laplace--Beltrami operator. We derive first order optimal potentials for the problem under consideration and find that the resulting solutions exhibit a surprising resemblance to the well-known Barenblatt--Prattle solution of the porous medium equation. Then, relying on these first order optimal potentials, we derive the pointwise $L^2$-limit of such discrete operators built from an i.i.d. random sample on a smooth compact manifold. Simulation results complementing the limiting distribution results are also presented.

Analysis of PDEs,Probability,Statistics Theory,Machine Learning
Conditions for existence of single valued optimal transport maps on convex boundaries with nontwisted cost

Seonghyeon Jeong,Jun Kitagawa

2023-08-14

Abstract:We prove that if $\Omega\subset \mathbb{R}^{n+1}$ is a (not necessarily strictly) convex, $C^1$ domain, and $\mu$ and $\bar{\mu}$ are probability measures absolutely continuous with respect to surface measure on $\partial \Omega$, with densities bounded away from zero and infinity, whose $2$-Monge-Kantorovich distance is sufficiently small, then there exists a continuous Monge solution to the optimal transport problem with cost function given by the quadratic distance on the ambient space $\mathbb{R}^{n+1}$. This result is also shown to be sharp, via a counterexample when $\Omega$ is uniformly convex but not $C^1$. Additionally, if $\Omega$ is $C^{1, \alpha}$ regular for some $\alpha$, then the Monge solution is shown to be Hölder continuous.

Analysis of PDEs
A geometric approach to apriori estimates for optimal transport maps

Simon Brendle,Flavien Léger,Robert J. McCann,Cale Rankin

2024-10-04

Abstract:A key inequality which underpins the regularity theory of optimal transport for costs satisfying the Ma--Trudinger--Wang condition is the Pogorelov second derivative bound. This translates to an apriori interior $C^1$ estimate for smooth optimal maps. Here we give a new derivation of this estimate which relies in part on Kim, McCann and Warren's observation that the graph of an optimal map becomes a volume maximizing spacelike submanifold when the product of the source and target domains is endowed with a suitable pseudo-Riemannian geometry that combines both the marginal densities and the cost.

Differential Geometry,Analysis of PDEs,Optimization and Control
Optimal transport map estimation in general function spaces

Vincent Divol,Jonathan Niles-Weed,Aram-Alexandre Pooladian

2024-01-03

Abstract:We study the problem of estimating a function $T$ given independent samples from a distribution $P$ and from the pushforward distribution $T_\sharp P$. This setting is motivated by applications in the sciences, where $T$ represents the evolution of a physical system over time, and in machine learning, where, for example, $T$ may represent a transformation learned by a deep neural network trained for a generative modeling task. To ensure identifiability, we assume that $T = \nabla \varphi_0$ is the gradient of a convex function, in which case $T$ is known as an \emph{optimal transport map}. Prior work has studied the estimation of $T$ under the assumption that it lies in a Hölder class, but general theory is lacking. We present a unified methodology for obtaining rates of estimation of optimal transport maps in general function spaces. Our assumptions are significantly weaker than those appearing in the literature: we require only that the source measure $P$ satisfy a Poincaré inequality and that the optimal map be the gradient of a smooth convex function that lies in a space whose metric entropy can be controlled. As a special case, we recover known estimation rates for Hölder transport maps, but also obtain nearly sharp results in many settings not covered by prior work. For example, we provide the first statistical rates of estimation when $P$ is the normal distribution and the transport map is given by an infinite-width shallow neural network.

Statistics Theory,Machine Learning
Gluing methods for quantitative stability of optimal transport maps

Cyril Letrouit,Quentin Mérigot

2024-11-08

Abstract:We establish quantitative stability bounds for the quadratic optimal transport map $T_\mu$ between a fixed probability density $\rho$ and a probability measure $\mu$ on $\mathbb{R}^d$. Under general assumptions on $\rho$, we prove that the map $\mu\mapsto T_\mu$ is bi-Hölder continuous, with dimension-free Hölder exponents. The linearized optimal transport metric $W_{2,\rho}(\mu,\nu)=\|T_\mu-T_\nu\|_{L^2(\rho)}$ is therefore bi-Hölder equivalent to the $2$-Wasserstein distance, which justifies its use in applications. We show this property in the following cases: (i) for any log-concave density $\rho$ with full support in $\mathbb{R}^d$, and any log-bounded perturbation thereof; (ii) for $\rho$ bounded away from $0$ and $+\infty$ on a John domain (e.g., on a bounded Lipschitz domain), while the only previously known result of this type assumed convexity of the domain; (iii) for some important families of probability densities on bounded domains which decay or blow-up polynomially near the boundary. Concerning the sharpness of point (ii), we also provide examples of non-John domains for which the Brenier potentials do not satisfy any Hölder stability estimate. Our proofs rely on local variance inequalities for the Brenier potentials in small convex subsets of the support of $\rho$, which are glued together to deduce a global variance inequality. This gluing argument is based on two different strategies of independent interest: one of them leverages the properties of the Whitney decomposition in bounded domains, the other one relies on spectral graph theory.

Analysis of PDEs,Probability
Optimal transport and regularity of weak Kantorovich potentials on a globally hyperbolic spacetime

Alec Metsch

2024-12-02

Abstract:We consider the optimal transportation problem on a globally hyperbolic spacetime for some Lorentzian cost function, which corresponds to the optimal transportation problem on a complete Riemannian manifold where the cost function is the Riemannian distance squared. We establish existence and uniqueness results for the optimal transport map and we investigate the regularity of weak Kantorovich potentials.

Optimization and Control,Mathematical Physics,Differential Geometry
Strict Convexity and C1,α Regularity of Potential Functions in Optimal Transportation under Condition A3w

Shibing Chen,Xu-Jia Wang

DOI: https://doi.org/10.1016/j.jde.2015.09.047

IF: 2.615

2016-01-01

Journal of Differential Equations

Abstract:In this paper we prove the strict c-convexity and the C1,α regularity for potential functions in optimal transportation under condition (A3w). These results were obtained by Caffarelli [1], [3], [4] for the cost c(x,y)=|x−y|2, by Liu [11], Loeper [15], Trudinger and Wang [20] for costs satisfying the condition (A3). For costs satisfying the condition (A3w), the results have also been proved by Figalli, Kim, and McCann [6], assuming that the initial and target domains are uniformly c-convex, see also [21]; and by Guillen and Kitagawa [8], assuming the cost function satisfies A3w in larger domains. In this paper we prove the strict c-convexity and the C1,α regularity assuming either the support of source density is compactly contained in a larger domain where the cost function satisfies A3w, or the dimension 2≤n≤4.
Partial Regularity for optimal transport maps

Guido De Philippis,Alessio Figalli

DOI: https://doi.org/10.48550/arXiv.1209.5640

2018-06-06

Abstract:We prove that, for general cost functions on $\mathbb{R}^n$, or for the cost $d^2/2$ on a Riemannian manifold, optimal transport maps between smooth densities are always smooth outside a closed singular set of measure zero.

Analysis of PDEs
Estimation of entropy-regularized optimal transport maps between non-compactly supported measures

Matthew Werenski,James M. Murphy,Shuchin Aeron

2023-11-21

Abstract:This paper addresses the problem of estimating entropy-regularized optimal transport (EOT) maps with squared-Euclidean cost between source and target measures that are subGaussian. In the case that the target measure is compactly supported or strongly log-concave, we show that for a recently proposed in-sample estimator, the expected squared $L^2$-error decays at least as fast as $O(n^{-1/3})$ where $n$ is the sample size. For the general subGaussian case we show that the expected $L^1$-error decays at least as fast as $O(n^{-1/6})$, and in both cases we have polynomial dependence on the regularization parameter. While these results are suboptimal compared to known results in the case of compactness of both the source and target measures (squared $L^2$-error converging at a rate $O(n^{-1})$) and for when the source is subGaussian while the target is compactly supported (squared $L^2$-error converging at a rate $O(n^{-1/2})$), their importance lie in eliminating the compact support requirements. The proof technique makes use of a bias-variance decomposition where the variance is controlled using standard concentration of measure results and the bias is handled by T1-transport inequalities along with sample complexity results in estimation of EOT cost under subGaussian assumptions. Our experimental results point to a looseness in controlling the variance terms and we conclude by posing several open problems.

Machine Learning,Information Theory,Statistics Theory
Entropic Optimal Transport: Convergence of Potentials

Marcel Nutz,Johannes Wiesel

DOI: https://doi.org/10.48550/arXiv.2104.11720

2021-10-30

Abstract:We study the potential functions that determine the optimal density for $\varepsilon$-entropically regularized optimal transport, the so-called Schrödinger potentials, and their convergence to the counterparts in classical optimal transport, the Kantorovich potentials. In the limit $\varepsilon\to0$ of vanishing regularization, strong compactness holds in $L^{1}$ and cluster points are Kantorovich potentials. In particular, the Schrödinger potentials converge in $L^{1}$ to the Kantorovich potentials as soon as the latter are unique. These results are proved for all continuous, integrable cost functions on Polish spaces. In the language of Schrödinger bridges, the limit corresponds to the small-noise regime.

Analysis of PDEs,Functional Analysis,Optimization and Control,Probability

Sobolev regularity for optimal transport maps of non-convex planar domains

Pointwise Schauder estimates for optimal transport maps of rough densities

Partial $W^{2,p}$ regularity for optimal transport maps

Global Regularity of Optimal Mappings in Non-Convex Domains

Central Limit Theorems for Smooth Optimal Transport Maps

On strict convexity and $C^1$ regularity of potential functions in optimal transportation

On optimal transport of matrix-valued measures

Sharp Sobolev inequalities on noncompact Riemannian manifolds with via optimal transport theory

Sharp Sobolev inequalities on noncompact Riemannian manifolds with ${\sf Ric}\geq 0$ via Optimal Transport theory

On Optimal Transport Maps Between 1 /d-Concave Densities

Weak limits of entropy regularized Optimal Transport; potentials, plans and divergences

Quadratically Regularized Optimal Transport: nearly optimal potentials and convergence of discrete Laplace operators

Conditions for existence of single valued optimal transport maps on convex boundaries with nontwisted cost

A geometric approach to apriori estimates for optimal transport maps

Optimal transport map estimation in general function spaces

Gluing methods for quantitative stability of optimal transport maps

Optimal transport and regularity of weak Kantorovich potentials on a globally hyperbolic spacetime

Strict Convexity and C1,α Regularity of Potential Functions in Optimal Transportation under Condition A3w

Partial Regularity for optimal transport maps

Estimation of entropy-regularized optimal transport maps between non-compactly supported measures

Entropic Optimal Transport: Convergence of Potentials