SHARP CONVERGENCE RATES FOR EMPIRICAL OPTIMAL TRANSPORT WITH SMOOTH COSTS

Tudor Manole, Jonathan Niles-Weed

Research output: Contribution to journalArticlepeer-review

Abstract

We revisit the question of characterizing the convergence rate of plug-in estimators of optimal transport costs. It is well known that an empirical measure comprising independent samples from an absolutely continuous distribution on Rd converges to that distribution at the rate n−1/d in Wasserstein distance, which can be used to prove that plug-in estimators of many optimal transport costs converge at this same rate. However, we show that when the cost is smooth, this analysis is loose: plug-in estimators based on empirical measures converge quadratically faster, at the rate n−2/d. As a corollary, we show that the Wasserstein distance between two distributions is significantly easier to estimate when the measures are well-separated. We also prove lower bounds, showing not only that our analysis of the plug-in estimator is tight, but also that no other estimator can enjoy significantly faster rates of convergence uniformly over all pairs of measures. Our proofs rely on empirical process theory arguments based on tight control of L2 covering numbers for locally Lipschitz and semiconcave functions. As a byproduct of our proofs, we derive L estimates on the displacement induced by the optimal coupling between any two measures satisfying suitable concentration and anticoncentration conditions, for a wide range of cost functions.

Original languageEnglish (US)
Pages (from-to)1108-1135
Number of pages28
JournalAnnals of Applied Probability
Volume34
Issue number1 B
DOIs
StatePublished - Feb 2024

Keywords

  • Optimal transport
  • Wasserstein distance
  • empirical measure
  • minimax bound

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Fingerprint

Dive into the research topics of 'SHARP CONVERGENCE RATES FOR EMPIRICAL OPTIMAL TRANSPORT WITH SMOOTH COSTS'. Together they form a unique fingerprint.

Cite this