Statistical bounds for entropic optimal transport: sample complexity and the central limit theorem
arXiv:1905.11882
Abstract
We prove several fundamental statistical bounds for entropic OT with the squared Euclidean cost between subgaussian probability measures in arbitrary dimension. First, through a new sample complexity result we establish the rate of convergence of entropic OT for empirical measures. Our analysis improves exponentially on the bound of Genevay et al. (2019) and extends their work to unbounded measures. Second, we establish a central limit theorem for entropic OT, based on techniques developed by Del Barrio and Loubes (2019). Previously, such a result was only known for finite metric spaces. As an application of our results, we develop and analyze a new technique for estimating the entropy of a random variable corrupted by gaussian noise.
Under review. 23 pages, 2 figures. Version 2 fixes minor typos and errors
References in corpus (5)
Cited by in corpus (12)
- Sinkhorn Divergences for Unbalanced Optimal Transport
- Entropic Optimal Transport between Unbalanced Gaussian Measures has a Closed Form
- Scalable Optimal Transport Methods in Machine Learning: A Contemporary Survey
- How Well Do WGANs Estimate the Wasserstein Metric?
- Debiased Sinkhorn barycenters
- Two-sample goodness-of-fit tests on the flat torus based on Wasserstein distance and their relevance to structural biology
- The statistical effect of entropic regularization in optimal transportation
- Gaussian-Smooth Optimal Transport: Metric Structure and Statistical Efficiency
- Two-sample Test with Kernel Projected Wasserstein Distance
- Uniform Confidence Band for Optimal Transport Map on One-Dimensional Data
- Estimating 2-Sinkhorn Divergence between Gaussian Processes from Finite-Dimensional Marginals
- Optimal transport weights for causal inference