2000 character limit reached
Convergence of Empirical Optimal Transport in Unbounded Settings (2306.11499v3)
Published 20 Jun 2023 in math.ST, math.PR, and stat.TH
Abstract: In compact settings, the convergence rate of the empirical optimal transport cost to its population value is well understood for a wide class of spaces and cost functions. In unbounded settings, however, hitherto available results require strong assumptions on the ground costs and the concentration of the involved measures. In this work, we pursue a decomposition-based approach to generalize the convergence rates found in compact spaces to unbounded settings under generic moment assumptions that are sharp up to an arbitrarily small $\epsilon > 0$. HaLLMark properties of empirical optimal transport on compact spaces, like the recently established adaptation to lower complexity, are shown to carry over to the unbounded case.
- Miklós Ajtai, János Komlós and Gábor Tusnády “On optimal matchings” In Combinatorica 4 Springer-Verlag, 1984, pp. 259–264
- Luigi Ambrosio, Michael Goldman and Dario Trevisan “On the quadratic random matching problem in two-dimensional domains” In Electronic Journal of Probability 27 The Institute of Mathematical Statisticsthe Bernoulli Society, 2022, pp. 1–35
- Luigi Ambrosio, Daniele Semola and Elia Brué “Lectures on Optimal Transport” Springer, 2021
- “Inequalities for the r𝑟ritalic_r-th absolute moment of a sum of random variables, 1≤r≤21𝑟21\leq r\leq 21 ≤ italic_r ≤ 2” In The Annals of Mathematical Statistics JSTOR, 1965, pp. 299–303
- “Combinatorial optimization over two random point sets” In Séminaire de Probabilités XLV Springer, 2013, pp. 483–535
- “One-dimensional empirical measures, order statistics, and Kantorovich transport distances” American Mathematical Society, 2019
- Sergey G Bobkov and Michel Ledoux “A simple Fourier analytic proof of the AKT optimal matching theorem” In The Annals of Applied Probability 31.6 Institute of Mathematical Statistics, 2021, pp. 2567–2584
- Vladimir Igorevich Bogachev “Measure theory” Springer, 2007
- Emmanuel Boissard and Thibaut Le Gouic “On the mean speed of convergence of empirical and occupation measures in Wasserstein distance” In Annales de l’IHP Probabilités et statistiques 50.2, 2014, pp. 539–563
- “Negative moments of positive random variables” In Journal of the American Statistical Association 67.338 Taylor & Francis, 1972, pp. 429–431
- “Faster Wasserstein distance estimation with the Sinkhorn divergence” In Advances in Neural Information Processing Systems 33, 2020, pp. 2257–2269
- Nabarun Deb, Promit Ghosal and Bodhisattva Sen “Rates of estimation of optimal transport maps using plug-in estimators via barycentric projections” In Advances in Neural Information Processing Systems 34 Curran Associates, Inc., 2021, pp. 29736–29753
- “Behavior of the empirical Wasserstein distance in ℝdsuperscriptℝ𝑑{\mathbb{R}}^{d}blackboard_R start_POSTSUPERSCRIPT italic_d end_POSTSUPERSCRIPT under moment conditions” In Electronic Journal of Probability 24.none Institute of Mathematical StatisticsBernoulli Society, 2019, pp. 1–32 DOI: 10.1214/19-EJP266
- Eustasio del Barrio, Alberto González-Sanz and Jean-Michel Loubes “Central limit theorems for semidiscrete Wasserstein distances” In Bernoulli [To Appear, preprint arXiv:2202.06380], 2023
- Steffen Dereich, Michael Scheutzow and Reik Schottstedt “Constructive quantization: Approximation by empirical measures” In Annales de l’IHP Probabilités et statistiques 49.4, 2013, pp. 1183–1203
- Vincent Divol “A short proof on the rate of convergence of the empirical measure for the Wasserstein distance” In Preprint arXiv:2101.08126, 2021
- V Dobrić and Joseph E Yukich “Asymptotics for transportation cost in high dimensions” In Journal of Theoretical Probability 8 Springer, 1995, pp. 97–118
- Richard Mansfield Dudley “The speed of mean Glivenko-Cantelli convergence” In The Annals of Mathematical Statistics 40.1 JSTOR, 1969, pp. 40–50
- “Statistical optimal transport via factored couplings” In Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics 89, Proceedings of Machine Learning Research PMLR, 2019, pp. 2454–2465 URL: https://proceedings.mlr.press/v89/forrow19a.html
- Nicolas Fournier “Convergence of the empirical measure in expected Wasserstein distance: Non asymptotic explicit bounds in ℝdsuperscriptℝ𝑑\mathbb{R}^{d}blackboard_R start_POSTSUPERSCRIPT italic_d end_POSTSUPERSCRIPT” In Preprint arXiv:2209.00923, 2022
- “On the rate of convergence in Wasserstein distance of the empirical measure” In Probability Theory and Related Fields 162.3 Springer, 2015, pp. 707–738
- A. Galichon “Optimal Transport Methods in Economics” Princeton University Press, 2016 URL: https://books.google.de/books?id=GDP9CwAAQBAJ
- “Convergence of asymptotic costs for random Euclidean matching problems” In Probability and Mathematical Physics 2.2 Mathematical Sciences Publishers, 2021, pp. 341–362
- “Improved training of Wasserstein GANs” In Advances in Neural Information Processing Systems 30, 2017
- “A unifying approach to distributional limits for empirical optimal transport” In Preprint arXiv:2202.12790, 2022
- Shayan Hundrieser, Thomas Staudt and Axel Munk “Empirical optimal transport between different measures adapts to lower complexity” In Annales de l’Institut Henri Poincaré [To Appear, preprint arXiv:2202.10434], 2023
- LV Kantorovich “On the translocation of masses” In Doklady Akademii Nauk URSS 37, 1942, pp. 7–8
- Benoı̂t R Kloeckner “Empirical measures: regularity is a counter-curse to dimensionality” In ESAIM: Probability and Statistics 24 EDP Sciences, 2020, pp. 408–434
- Nicholas Kolkin, Jason Salavon and Gregory Shakhnarovich “Style transfer by relaxed optimal transport and self-similarity” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 10051–10060
- “ϵitalic-ϵ\epsilonitalic_ϵ-Entropy and ϵitalic-ϵ\epsilonitalic_ϵ-Capacity of sets in functional spaces” In Twelve Papers on Algebra and Real Functions, American Mathematical Society Translations–series 2 American Mathematical Society, 1961, pp. 277–364
- Martin Larsson, Jonghwa Park and Johannes Wiesel “On concentration of the empirical measure for general transport costs” In Preprint arXiv:2305.18636, 2023
- M Ledoux “On optimal matching of Gaussian samples II” In Preprint, 2018
- M Ledoux “On optimal matching of Gaussian samples” In Journal of Mathematical Sciences 238.4, 2019, pp. 495–522
- “On optimal matching of Gaussian samples III” In Probability and Mathematical Statistics 41.2, 2021, pp. 237–265
- Jing Lei “Convergence and concentration of empirical measures under Wasserstein distance in unbounded functional spaces” In Bernoulli 26.1, 2020, pp. 767–798
- “Plugin estimation of smooth optimal transport maps” In Preprint arXiv:2107.12364, 2021
- “Sharp convergence rates for empirical optimal transport with smooth costs” In The Annals of Applied Probability [To appear, preprint arXiv:2106.13181], 2023
- Gaspard Monge “Mémoire sur la théorie des déblais et des remblais” In Histoire de l’Académie Royale des Sciences de Paris, 1781, pp. 666–704
- Alfred Müller “Integral probability metrics and their generating classes of functions” In Advances in applied probability 29.2 Cambridge University Press, 1997, pp. 429–443
- “Estimation of Wasserstein distances in the spiked transport model” In Bernoulli 28.4 Bernoulli Society for Mathematical StatisticsProbability, 2022, pp. 2663–2688
- John P Nolan “Univariate Stable Distributions: Models for Heavy Tailed Data” In Springer Series in Operations Research and Financial Engineering Springer, 2020
- Victor M Panaretos and Yoav Zemel “An Invitation to Statistics in Wasserstein Space” Springer Nature, 2020
- “Computational Optimal Transport: With Applications to Data Science” In Foundations and Trends® in Machine Learning 11.5-6 Now Publishers, Inc., 2019, pp. 355–607
- F. Santambrogio “Optimal Transport for Applied Mathematicians: Calculus of Variations, PDEs, and Modeling”, Progress in Nonlinear Differential Equations and Their Applications Springer, 2015
- “Optimal-transport analysis of single-cell gene expression identifies developmental trajectories in reprogramming” In Cell 176.4, 2019, pp. 928–943.e22 DOI: https://doi.org/10.1016/j.cell.2019.01.006
- “Minimax distribution estimation in Wasserstein distance” In Preprint arXiv:1802.08855, 2018
- “Inference for empirical Wasserstein distances on finite spaces” In Journal of the Royal Statistical Society: Series B (Statistical Methodology) 80.1 Wiley Online Library, 2018, pp. 219–238
- Thomas Staudt, Shayan Hundrieser and Axel Munk “On the uniqueness of Kantorovich potentials” In Preprint arXiv:2201.08316, 2022
- Michel Talagrand “Matching theorems and empirical discrepancy computations using majorizing measures” In Journal of the American Mathematical Society 7.2, 1994, pp. 455–537
- Michel Talagrand “The transportation cost from the uniform measure to the empirical measure in dimension ≥\geq≥ 3” In The Annals of Probability JSTOR, 1994, pp. 919–959
- “Colocalization for super-resolution microscopy via optimal transport” In Nature Computational Science 1.3 Nature Publishing Group, 2021, pp. 199–211
- Aad W. Vaart and John A. Wellner “Weak Convergence and Empirical Processes: With Applications to Statistics”, Springer Series in Statistics Springer, 1996
- C. Villani “Optimal Transport: Old and New”, A Series of Comprehensive Studies in Mathematics Springer, 2008
- “Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance” In Bernoulli 25.4A Bernoulli Society for Mathematical StatisticsProbability, 2019, pp. 2620–2648
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.
