Minimax-Optimal Two-Sample Test with Sliced Wasserstein

TL;DR

This paper introduces a permutation-based two-sample test using sliced Wasserstein distance, achieving minimax optimal separation rates with strong finite-sample guarantees and practical scalability.

Contribution

It develops the first theoretical analysis of a sliced Wasserstein-based test, establishing minimax optimality and finite-sample error control, with practical advantages over kernel methods.

Findings

Achieves $n^{-1/2}$ minimax separation rate.

Maintains finite-sample Type I error control.

Demonstrates competitive power and scalability in experiments.

Abstract

We study the problem of nonparametric two-sample testing using the sliced Wasserstein (SW) distance. While prior theoretical and empirical work indicates that the SW distance offers a promising balance between strong statistical guarantees and computational efficiency, its theoretical foundations for hypothesis testing remain limited. We address this gap by proposing a permutation-based SW test and analyzing its performance. The test inherits finite-sample Type I error control from the permutation principle. Moreover, we establish non-asymptotic power bounds and show that the procedure achieves the minimax separation rate $n^{-1/2}$ over multinomial and bounded-support alternatives, matching the optimal guarantees of kernel-based tests while building on the geometric foundations of Wasserstein distances. Our analysis further quantifies the trade-off between the number of projections and statistical power. Finally, numerical experiments demonstrate that the test combines finite-sample validity with competitive power and scalability, and -- unlike kernel-based tests, which require careful kernel tuning -- it performs consistently well across all scenarios we consider.