Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance

DOI10.3150/18-BEJ1065zbMath1428.62099arXiv1707.00087MaRDI QIDQ2325336

Jonathan Weed, Francis Bach

Publication date: 25 September 2019

Published in: Bernoulli (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1707.00087

zbMATH Keywords

quantizationoptimal transportWasserstein metrics

Mathematics Subject Classification ID

Inequalities; stochastic orderings (60E15) Applications of statistics to physics (62P35) Convergence of probability measures (60B10)

Related Items

GAT–GMM: Generative Adversarial Training for Gaussian Mixture Models, Regularization via Mass Transportation, Convergence and finite sample approximations of entropic regularized Wasserstein distances in Gaussian and RKHS settings, Applications of optimal transportation in the natural sciences. Abstracts from the workshop held February 21--27, 2021 (online meeting), Multivariate ranks and quantiles using optimal transport: consistency, rates and nonparametric testing, Minimax confidence intervals for the sliced Wasserstein distance, Measure estimation on manifolds: an optimal transport approach, Stochastic approximation versus sample average approximation for Wasserstein barycenters, Minimax estimation of smooth densities in Wasserstein distance, Sparse optimization on measures with over-parameterized gradient descent, Randomized Wasserstein Barycenter Computation: Resampling with Statistical Guarantees, Unnamed Item, Entropy-regularized 2-Wasserstein distance between Gaussian measures, Reweighting samples under covariate shift using a Wasserstein distance criterion, Deep Unfitted Nitsche Method for Elliptic Interface Problems, Berry-Esseen smoothing inequality for the Wasserstein metric on compact Lie groups, Convergence rates for empirical measures of Markov chains in dual and Wasserstein distances, $k$-Variance: A Clustered Notion of Variance, Estimation of Wasserstein distances in the spiked transport model, Measuring association with Wasserstein distances, On the capacity of deep generative networks for approximating distributions, A class of dimension-free metrics for the convergence of empirical measures, On adaptive confidence sets for the Wasserstein distances, Sharp convergence rates for empirical optimal transport with smooth costs, A blob method for inhomogeneous diffusion with applications to multi-agent control and sampling, An Improved Central Limit Theorem and Fast Convergence Rates for Entropic Transportation Costs, Empirical measures and random walks on compact spaces in the quadratic Wasserstein metric, Two-sample goodness-of-fit tests on the flat torus based on Wasserstein distance and their relevance to structural biology, Central limit theorems for semi-discrete Wasserstein distances, Control of neural transport for normalising flows, Asymptotics for Strassen's optimal transport problem, Quantitative stability of barycenters in the Wasserstein space, Limit theorems for entropic optimal transport maps and Sinkhorn divergence, An optimal transport approach for the Schrödinger bridge problem and convergence of Sinkhorn algorithm, Behavior of the empirical Wasserstein distance in \({\mathbb R}^d\) under moment conditions, Empirical measures: regularity is a counter-curse to dimensionality, Scaling Limit of the Stein Variational Gradient Descent: The Mean Field Regime, A non-exponential extension of Sanov’s theorem via convex duality, Posterior asymptotics in Wasserstein metrics on the real line, Bounding the expectation of the supremum of empirical processes indexed by Hölder classes, Wasserstein upper bounds of the total variation for smooth densities, Convergence and concentration of empirical measures under Wasserstein distance in unbounded functional spaces, Hybrid Wasserstein distance and fast distribution clustering, On a Kantorovich-Rubinstein inequality, Empirical optimal transport on countable metric spaces: distributional limits and statistical applications, Unnamed Item, Minimax estimation of smooth optimal transport maps, Multivariate goodness-of-fit tests based on Wasserstein distance, A topological study of functional data and Fréchet functions of metric measure spaces, Generalization error of GAN from the discriminator's perspective, Surrogate parametric metamodel based on optimal transport, Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance, Uniform rates of the Glivenko-Cantelli convergence and their use in approximating Bayesian inferences, Finer estimates on the \(2\)-dimensional matching problem, Dynamics of Data-driven Ambiguity Sets for Hyperbolic Conservation Laws with Uncertain Inputs, Optimal Transport to a Variety, Finite Sample Approximations of Exact and Entropic Wasserstein Distances Between Covariance Operators and Gaussian Processes, Limitations of the Wasserstein MDE for univariate data, A measure theoretical approach to the mean-field maximum principle for training NeurODEs, Hybrid twins based on optimal transport, Likelihood estimation of sparse topic distributions in topic models and its applications to Wasserstein document distance calculations, Learning Finite-Dimensional Coding Schemes with Nonlinear Reconstruction Maps

Uses Software

• ElemStatLearn
• EMD

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Constructive quantization: approximation by empirical measures
• Long history of the Monge-Kantorovich transportation problem
• Simple bounds for the convergence of empirical and occupation measures in 1-Wasserstein distance
• On the rate of convergence in Wasserstein distance of the empirical measure
• Sublinear time algorithms for earth mover's distance
• Matrix random products with singular harmonic measure
• Quantitative concentration inequalities for empirical measures on non-compact spaces
• Morceaux de graphes lipschitziens et intégrales singulières sur une surface
• Isoperimetric and concentration inequalities: equivalence under curvature lower bound
• On optimal matchings
• Some relations between dimension and Lyapunov exponents
• Matching random samples in many dimensions
• Deterministic and stochastic error bounds in numerical analysis
• A characterization of random variables with minimum \(L^ 2\)-distance
• Exponential integrability and transportation cost related to logarithmic Sobolev inequalities
• Asymptotics for transportation cost in high dimensions
• A space quantization method for numerical integration
• The earth mover's distance as a metric for image retrieval
• Introduction to rare event simulation.
• Remarks on logarithmic Sobolev constants, exponential integrability and bounds on the diameter
• Concentration of measure and isoperimetric inequalities in product spaces
• A sharp estimate of the binomial mean absolute deviation with applications
• Foundations of quantization for probability distributions
• Functional quantization-based stratified sampling methods
• Sharp asymptotic and finite-sample rates of convergence of empirical measures in Wasserstein distance
• Optimal transport for applied mathematicians. Calculus of variations, PDEs, and modeling
• Multiscale geometric methods for data sets. I: Multiscale SVD, noise and curvature.
• On the mean speed of convergence of empirical and occupation measures in Wasserstein distance
• Convolutional wasserstein distances
• Approximation by finitely supported measures
• Dimension, entropy and Lyapunov exponents
• Elements of Information Theory
• Epsilon Entropy of Stochastic Processes
• The Speed of Mean Glivenko-Cantelli Convergence
• Convergence of Entropic Schemes for Optimal Transport and Gradient Flows
• Convergence de la répartition empirique vers la répartition théorique
• Optimal Transport
• The elements of statistical learning. Data mining, inference, and prediction