OPTIMAL TRANSPORT MAPS

. We study statistical inference for the optimal transport (OT) map (also known as the Brenier map) from a known absolutely continuous reference distribution onto an unknown finitely discrete target distribution. We derive limit distributions for the L p -estimation error with arbitrary p ∈ [1 , ∞ ) and for linear functionals of the empirical OT map. The former has a non-Gaussian limit, while the latter attains asymptotic normality. For both cases, we also establish consistency of the nonparametric bootstrap. The derivation of our limit theorems relies on new stability estimates of functionals of the OT map with respect to the dual potential vector, which could be of independent interest.

[1]  Jonathan Niles-Weed,et al.  Minimax estimation of discontinuous optimal transport maps: The semi-discrete case , 2023, ICML.

[2]  Jonathan Niles-Weed,et al.  Optimal transport map estimation in general function spaces , 2022, 2212.03722.

[3]  Kengo Kato,et al.  Limit Theorems for Entropic Optimal Transport Maps and the Sinkhorn Divergence , 2022, 2207.08683.

[4]  Jean-Michel Loubes,et al.  Weak limits of entropy regularized Optimal Transport; potentials, plans and divergences , 2022, 2207.07427.

[5]  Austin J. Stromme,et al.  On the sample complexity of entropic optimal transport , 2022, 2206.13472.

[6]  G. Carlier,et al.  Convergence rate of general entropic optimal transport costs , 2022, Calculus of Variations and Partial Differential Equations.

[7]  Kengo Kato,et al.  Limit distribution theory for smooth $p$-Wasserstein distances , 2022, 2203.00159.

[8]  Marco Cuturi,et al.  Debiaser Beware: Pitfalls of Centering Regularized Transport Maps , 2022, ICML.

[9]  Jonathan Niles-Weed,et al.  Entropic estimation of optimal transport maps , 2021, 2109.12004.

[10]  Florian Gunsilius,et al.  An Optimal Transport Approach to Causal Inference , 2021, 2108.05858.

[11]  L. Wasserman,et al.  Plugin Estimation of Smooth Optimal Transport Maps , 2021, 2107.12364.

[12]  Bodhisattva Sen,et al.  Rates of Estimation of Optimal Transport Maps using Plug-in Estimators via Barycentric Projections , 2021, NeurIPS.

[13]  Jean-Michel Loubes,et al.  A Central Limit Theorem for Semidiscrete Wasserstein Distances , 2021, 2202.06380.

[14]  Carlos Matrán,et al.  Distribution and quantile functions, ranks and signs in dimension d: A measure transportation approach , 2021, The Annals of Statistics.

[15]  Philippe Rigollet,et al.  Minimax estimation of smooth optimal transport maps , 2021 .

[16]  Yoav Zemel,et al.  An Invitation to Statistics in Wasserstein Space , 2020 .

[17]  J. Kitagawa,et al.  Quantitative Stability in the Geometry of Semi-discrete Optimal Transport , 2020, International Mathematics Research Notices.

[18]  Jonathan Niles-Weed,et al.  Estimation of Wasserstein distances in the Spiked Transport Model , 2019, Bernoulli.

[19]  Marco Cuturi,et al.  Computational Optimal Transport: With Applications to Data Science , 2019 .

[20]  B. Sen,et al.  Multivariate ranks and quantiles using optimal transport: Consistency, rates and nonparametric testing , 2019, The Annals of Statistics.

[21]  Kengo Kato,et al.  Detailed proof of Nazarov's inequality , 2017, 1711.10696.

[22]  Bruno Lévy,et al.  Notions of optimal transport theory and how to implement them on a computer , 2017, Comput. Graph..

[23]  F. Bach,et al.  Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance , 2017, Bernoulli.

[24]  F. Santambrogio.  {Euclidean, metric, and Wasserstein} gradient flows: an overview , 2016, 1609.03890.

[25]  Victor Chernozhukov,et al.  Generic Inference on Quantile and Quantile Effect Functions for Discrete Outcomes , 2016, Journal of the American Statistical Association.

[26]  Q. Mérigot,et al.  A Newton algorithm for semi-discrete optimal transport with storage fees and quantitative convergence of cells , 2019, SIAM J. Optim..

[27]  V. Chernozhukov,et al.  Monge-Kantorovich Depth, Quantiles, Ranks and Signs , 2014, 1412.8434.

[28]  G. Carlier,et al.  Vector Quantile Regression , 2014, 1406.4643.

[29]  Andrés Santos,et al.  Inference on Directionally Differentiable Functions , 2014, The Review of Economic Studies.

[30]  A. Guillin,et al.  On the rate of convergence in Wasserstein distance of the empirical measure , 2013, 1312.2128.

[31]  Marco Cuturi,et al.  Sinkhorn Distances: Lightspeed Computation of Optimal Transport , 2013, NIPS.

[32]  Quentin Mérigot,et al.  A Multiscale Approach to Optimal Transport , 2011, Comput. Graph. Forum.

[33]  C. Villani.  Optimal Transport: Old and New , 2008 .

[34]  E. Milman.  On the role of convexity in isoperimetry, spectral gap and concentration , 2007, 0712.4092.

[35]  Stergios B. Fotopoulos,et al.  All of Nonparametric Statistics , 2007, Technometrics.

[36]  Werner Römisch,et al.  Delta Method, Infinite Dimensional , 2006 .

[37]  T. Mikami.  Monge’s problem with a quadratic cost by the zero-noise limit of h-path processes , 2004 .

[38]  Jon A. Wellner,et al.  Weak Convergence and Empirical Processes: With Applications to Statistics , 1996 .

[39]  Y. Brenier.  Polar Factorization and Monotone Rearrangement of Vector-Valued Functions , 1991 .

[40]  Franz Aurenhammer,et al.  Power Diagrams: Properties, Algorithms and Applications , 1987, SIAM J. Comput..

[41]  R. Dudley.  The Speed of Mean Glivenko-Cantelli Convergence , 1969 .

[42]  Filippo Santambrogio,et al.  Optimal Transport for Applied Mathematicians , 2015 .

[43]  G. Saridis,et al.  Journal of Optimization Theory and Applications Approximate Solutions to the Time-invariant Hamilton-jacobi-bellman Equation 1 , 1998 .

[44]  L. Ambrosio,et al.  Gradient Flows: In Metric Spaces and in the Space of Probability Measures , 2005 .

[45]  O. Mangasarian.  On Concepts of Directional Differentiability , 2004 .

[46]  Sophie Lambert-Lacroix,et al.  On nonparametric confidence set estimation , 2001 .

[47]  S. Bobkov.  Isoperimetric and Analytic Inequalities for Log-Concave Probability Measures , 1999 .

[48]  Franz Aurenhammer,et al.  Minkowski-Type Theorems and Least-Squares Clustering , 1998, Algorithmica.

[49]  Lutz Diimbgen.  On nondifferentiable functions and the bootstrap , 1993 .

[50]  L. Evans.  Measure theory and fine properties of functions , 1992 .

[51]  Adrian Bowyer,et al.  Computing Dirichlet Tessellations , 1981, Comput. J..

[52]  D. F. Watson.  Computing the n-Dimensional Delaunay Tesselation with Application to Voronoi Polytopes , 1981, Comput. J..