Maximum Likelihood Estimates for Gaussian Mixtures Are Transcendental

DOI10.1007/978-3-319-32859-1_49zbMATH Open1460.62081OpenAlexW1953619313WikidataQ57566381 ScholiaQ57566381MaRDI QIDQ2807877FDOQ2807877

Authors Mathias Drton, Bernd Sturmfels, Carlos Enrique Améndola Cerón 

Publication date 25 May 2016

Published in Mathematical Aspects of Computer and Information Sciences (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1508.06958

zbMATH Keywords

expectation maximizationmaximum likelihoodmixture modelnormal distributionalgebraic statisticstranscendence theory

Mathematics Subject Classification ID

Point estimation (62F10) Learning and adaptive systems in artificial intelligence (68T05) Estimation in multivariate analysis (62H12)

Abstract: Gaussian mixture models are central to classical statistics, widely used in the information sciences, and have a rich mathematical structure. We examine their maximum likelihood estimates through the lens of algebraic statistics. The MLE is not an algebraic function of the data, so there is no notion of ML degree for these models. The critical points of the likelihood function are transcendental, and there is no bound on their number, even for mixtures of two univariate Gaussians.

Recommendations

• Maximum likelihood estimation of heterogeneous mixtures of Gaussian and uniform distributions
• MAXIMUM LIKELIHOOD PARAMETER ESTIMATION FOR THE MIXTURES OF NORMAL DISTRIBUTIONS
• Maximum likelihood estimation for \(N\)-mixture models
• Maximum likelihood estimation of Gaussian mixture models without matrix operations
• A note on maximum likelihood estimation for mixture models
• Generalized maximum likelihood estimation of normal mixture densities
• scientific article; zbMATH DE number 4180561
• A Comment on Maximum Likelihood Estimation for Finite Mixtures of Distributions
• Maximum likelihood estimators for normal and gamma mixtures

Cited in

(4)

• Exact solutions in log-concave maximum likelihood estimation
• Maximum number of modes of Gaussian mixtures
• Identifying the number of components in Gaussian mixture models using numerical algebraic geometry
• Injectivity of real rational mappings: the case of a mixture of two Gaussian laws

Describes a project that uses

Uses Software

• mclust

This page was built for publication: Maximum Likelihood Estimates for Gaussian Mixtures Are Transcendental

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2807877)