Characteristic and universal tensor product kernels
From MaRDI portal
Abstract: Maximum mean discrepancy (MMD), also called energy distance or N-distance in statistics and Hilbert-Schmidt independence criterion (HSIC), specifically distance covariance in statistics, are among the most popular and successful approaches to quantify the difference and independence of random variables, respectively. Thanks to their kernel-based foundations, MMD and HSIC are applicable on a wide variety of domains. Despite their tremendous success, quite little is known about when HSIC characterizes independence and when MMD with tensor product kernel can discriminate probability distributions. In this paper, we answer these questions by studying various notions of characteristic property of the tensor product kernel.
Recommendations
- Equivalence of distance-based and RKHS-based statistics in hypothesis testing
- Kernel distribution embeddings: universal kernels, characteristic kernels and kernel metrics on distributions
- Multivariate tests of independence based on a new class of measures of independence in reproducing kernel Hilbert space
- Two-sample statistics based on anisotropic kernels
- Algorithmic Learning Theory
Cites work
- A characterization of distributions by mean values of statistics and certain probabilistic metrics
- A Hilbert Space Embedding for Distributions
- A kernel two-sample test
- A new test for multivariate normality
- Algorithmic Learning Theory
- Brownian distance covariance
- Computing functions of random variables via reproducing kernel Hilbert space representations
- Distance covariance in metric spaces
- Distinguishing cause from effect using observational data: methods and benchmarks
- Equivalence of distance-based and RKHS-based statistics in hypothesis testing
- Feature selection via dependence maximization
- Hilbert space embeddings and metrics on probability measures
- scientific article; zbMATH DE number 5055767 (Why is no real title available?)
- Kernel Bayes' rule: Bayesian inference with positive definite kernels
- Kernel dimension reduction in regression
- Kernel Mean Embedding of Distributions: A Review and Beyond
- Kernel methods for measuring independence
- Kernel-based tests for joint independence
- Learning theory for distribution regression
- Measuring and testing dependence by correlation of distances
- On a new multivariate two-sample test.
- On the influence of the kernel on the consistency of support vector machines
- Scattered Data Approximation
- Support Vector Machines
- Theory of Reproducing Kernels
- Universal kernels
- Universality, Characteristic Kernels and RKHS Embedding of Measures
- VECTOR VALUED REPRODUCING KERNEL HILBERT SPACES AND UNIVERSALITY
Cited in
(15)- Learning rate of distribution regression with dependent samples
- Minimum-energy measures for singular kernels
- Multivariate tests of independence based on a new class of measures of independence in reproducing kernel Hilbert space
- Bayesian quadrature, energy minimization, and space-filling design
- Universality, Characteristic Kernels and RKHS Embedding of Measures
- Generalization of the HSIC and distance covariance using PDI kernels
- Consistency of permutation tests of independence using distance covariance, HSIC and dHSIC
- Kernel-based Sensitivity Analysis for (Excursion) Sets
- A nonparametric test for elliptical distribution based on kernel embedding of probabilities
- Robust Bayesian inference for measurement error misspecification: the Berkson and classical cases
- Doubly robust conditional independence testing with generative neural networks
- Distribution free tests for model selection based on maximum mean discrepancy with estimated parameters
- Kernel Angle Dependence Measures in Metric Spaces
- Global Sensitivity Analysis of Spatial Sets: Finite-Element-Based Estimation and the Role of Observation Windows
- Equivalence of distance-based and RKHS-based statistics in hypothesis testing
This page was built for publication: Characteristic and universal tensor product kernels
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4558575)
