Asymptotically Optimal One- and Two-Sample Testing With Kernels

DOI10.1109/TIT.2021.3059267MaRDI QIDQ5001613zbMATH OpenOpenAlexFDO

Authors Shengyu Zhu, Biao Chen, Zhitang Chen, Peng-fei Yang

Publication date 22 July 2021

Published in IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1908.10037

zbMATH Keywords

extended version of Sanov's theoremmaximum mean discrepancy (MMD) based test

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10)

Abstract: We characterize the asymptotic performance of nonparametric one- and two-sample testing. The exponential decay rate or error exponent of the type-II error probability is used as the asymptotic performance metric, and an optimal test achieves the maximum rate subject to a constant level constraint on the type-I error probability. With Sanov's theorem, we derive a sufficient condition for one-sample tests to achieve the optimal error exponent in the universal setting, i.e., for any distribution defining the alternative hypothesis. We then show that two classes of Maximum Mean Discrepancy (MMD) based tests attain the optimal type-II error exponent on $mathbb{R}^d$, while the quadratic-time Kernel Stein Discrepancy (KSD) based tests achieve this optimality with an asymptotic level constraint. For general two-sample testing, however, Sanov's theorem is insufficient to obtain a similar sufficient condition. We proceed to establish an extended version of Sanov's theorem and derive an exact error exponent for the quadratic-time MMD based two-sample tests. The obtained error exponent is further shown to be optimal among all two-sample tests satisfying a given level constraint. Our work hence provides an achievability result for optimal nonparametric one- and two-sample testing in the universal setting. Application to off-line change detection and related issues are also discussed.

Recommendations

• On Asymptotic Minimaxity of Kernel-based Tests
• A kernel two-sample test
• Bayesian Kernel Two-Sample Testing
• Asymptotics and practical aspects of testing normality with kernel methods
• A new distribution-free \(k\)-sample test: analysis of kernel density functionals
• A reproducing kernel Hilbert space log‐rank test for the two‐sample problem
• Kernel two-sample tests in high dimensions: interplay between moment discrepancy and dimension-and-sample orders
• A one-sample test for normality with kernel methods
• scientific article; zbMATH DE number 3915425
• \(k\)-sample test based on the common area of kernel density estimators

Cited in

(4)

• Large-scale simultaneous testing using kernel density estimation
• On Asymptotic Minimaxity of Kernel-based Tests
• New bounds for the empirical robust Kullback-Leibler divergence problem
• Kullback-Leibler divergence based multidimensional robust universal hypothesis testing

This page was built for publication: Asymptotically Optimal One- and Two-Sample Testing With Kernels

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5001613)