Learning Gaussian mixtures using the Wasserstein-Fisher-Rao gradient flow

DOI10.1214/24-AOS2416MaRDI QIDQ6621546zbMATH OpenFDO

Authors Yuling Yan, Kaizheng Wang, Philippe Rigollet

Publication date 18 October 2024

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1214/24-AOS2416

zbMATH Keywords

Gaussian mixture modelnonparametric MLEoptimal transportoverparameterizationWasserstein gradient flowsWasserstein-Fisher-Rao geometry

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Cites work

• \{Euclidean, metric, and Wasserstein\} gradient flows: an overview
• A JKO splitting scheme for Kantorovich-Fisher-Rao gradient flows
• A mean field view of the landscape of two-layer neural networks
• A new optimal transport distance on the space of finite Radon measures
• An interpolating distance between optimal transport and Fisher-Rao metrics
• Birth–death dynamics for sampling: global convergence, approximations and their asymptotics
• Consistency of the Maximum Likelihood Estimator in the Presence of Infinitely Many Incidental Parameters
• Convex Optimization, Shape Constraints, Compound Decisions, and Empirical Bayes Rules
• General maximum likelihood empirical Bayes estimation of normal means
• Generalized maximum likelihood estimation of normal mixture densities
• Gradient flows in metric spaces and in the space of probability measures
• High-dimensional classification via nonparametric empirical Bayes and maximum likelihood inference
• Mixtures of exponential distributions
• On the nonparametric maximum likelihood estimator for Gaussian location mixture densities with application to Gaussian denoising
• Optimal entropy-transport problems and a new Hellinger-Kantorovich distance between positive measures
• Optimal Transport
• THE GEOMETRY OF DISSIPATIVE EVOLUTION EQUATIONS: THE POROUS MEDIUM EQUATION
• The geometry of mixture likelihoods: A general theory
• The Support Reduction Algorithm for Computing Non‐Parametric Function Estimates in Mixture Models
• The Variational Formulation of the Fokker--Planck Equation
• Uniqueness of estimation and identifiability in mixture models
• Uniqueness of the Fisher-Rao metric on the space of smooth densities

Cited in

(4)

• Optimal Transport based Cross-Domain Integration for Heterogeneous Data
• Nonparametric empirical Bayes prediction in mixed models
• A particle algorithm for mean-field variational inference
• Two statistical problems for multivariate mixture distributions

This page was built for publication: Learning Gaussian mixtures using the Wasserstein-Fisher-Rao gradient flow

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6621546)