Parsimonious Gaussian mixture models with piecewise-constant eigenvalue profiles

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Szwagier, Tom, Mattei, Pierre-Alexandre, Bouveyron, Charles, Pennec, Xavier
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912692529266688
author Szwagier, Tom
Mattei, Pierre-Alexandre
Bouveyron, Charles
Pennec, Xavier
author_facet Szwagier, Tom
Mattei, Pierre-Alexandre
Bouveyron, Charles
Pennec, Xavier
contents Gaussian mixture models (GMMs) are ubiquitous in statistical learning, particularly for unsupervised problems. While full GMMs suffer from the overparameterization of their covariance matrices in high-dimensional spaces, spherical GMMs (with isotropic covariance matrices) certainly lack flexibility to fit certain anisotropic distributions. Connecting these two extremes, we introduce a new family of parsimonious GMMs with piecewise-constant covariance eigenvalue profiles. These extend several low-rank models like the celebrated mixtures of probabilistic principal component analyzers (MPPCA), by enabling any possible sequence of eigenvalue multiplicities. If the latter are prespecified, then we can naturally derive an expectation-maximization (EM) algorithm to learn the mixture parameters. Otherwise, to address the notoriously-challenging issue of jointly learning the mixture parameters and hyperparameters, we propose a componentwise penalized EM algorithm, whose monotonicity is proven. We show the superior likelihood-parsimony tradeoffs achieved by our models on a variety of unsupervised experiments: density fitting, clustering and single-image denoising.
format Preprint
id arxiv_https___arxiv_org_abs_2507_01542
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Parsimonious Gaussian mixture models with piecewise-constant eigenvalue profiles
Szwagier, Tom
Mattei, Pierre-Alexandre
Bouveyron, Charles
Pennec, Xavier
Machine Learning
Applications
Computation
Methodology
Gaussian mixture models (GMMs) are ubiquitous in statistical learning, particularly for unsupervised problems. While full GMMs suffer from the overparameterization of their covariance matrices in high-dimensional spaces, spherical GMMs (with isotropic covariance matrices) certainly lack flexibility to fit certain anisotropic distributions. Connecting these two extremes, we introduce a new family of parsimonious GMMs with piecewise-constant covariance eigenvalue profiles. These extend several low-rank models like the celebrated mixtures of probabilistic principal component analyzers (MPPCA), by enabling any possible sequence of eigenvalue multiplicities. If the latter are prespecified, then we can naturally derive an expectation-maximization (EM) algorithm to learn the mixture parameters. Otherwise, to address the notoriously-challenging issue of jointly learning the mixture parameters and hyperparameters, we propose a componentwise penalized EM algorithm, whose monotonicity is proven. We show the superior likelihood-parsimony tradeoffs achieved by our models on a variety of unsupervised experiments: density fitting, clustering and single-image denoising.
title Parsimonious Gaussian mixture models with piecewise-constant eigenvalue profiles
topic Machine Learning
Applications
Computation
Methodology
url https://arxiv.org/abs/2507.01542