Hierarchical low-rank structure of parameterized distributions

DOI10.4310/CMS.2021.V19.N3.A14MaRDI QIDQ2057039zbMATH OpenFDO

Authors Jun Qin, Lexing Ying

Publication date 8 December 2021

Published in Communications in Mathematical Sciences (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1911.13277

zbMATH Keywords

hierarchical low-ranknessKullback-Leibler (KL) divergenceparameterized distributions

Mathematics Subject Classification ID

Approximation by operators (in particular, by integral operators) (41A35) Approximations to statistical distributions (nonasymptotic) (62E17) Multivariate distribution of statistics (62H10) Information theory (general) (94A15)

Abstract: This note shows that the matrix forms of several one-parameter distribution families satisfy a hierarchical low-rank structure. Such families of distributions include binomial, Poisson, and $ch{i}^2$ distributions. The proof is based on a uniform relative bound of a related divergence function. Numerical results are provided to confirm the theoretical findings.

Recommendations

• A Note on the One Parameter Power Series Distribution
• Probability matrices, non-negative rank, and parameterization of mixture models
• Logarithmically complete monotonicity of a matrix-parametrized analogue of the multinomial distribution
• Multivariate log-concave distributions as a nearly parametric model
• scientific article; zbMATH DE number 1420776

Cited in

(1)

• Normal approximation for hierarchical structures

This page was built for publication: Hierarchical low-rank structure of parameterized distributions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2057039)