Parsimonious Gaussian mixture models with piecewise-constant eigenvalue profiles
Preprint
Abstract
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.Keywords
All Related Versions
This publication has 61 references indexed in Scilit:
- Eigengap Sparsity for Covariance Parsimony2025
- Structured prior distributions for the covariance matrix in latent factor modelsStatistics and Computing, 2024
- Parsimonious ultrametric Gaussian mixture modelsStatistics and Computing, 2024
- Optimization on flag manifoldsMathematical Programming, 2021
- An alternative to EM for Gaussian mixture models: batch and stochastic Riemannian optimizationMathematical Programming, 2019
- Model-based clustering based on sparse finite Gaussian mixturesStatistics and Computing, 2014
- Parsimonious Gaussian mixture modelsStatistics and Computing, 2008
- Principal Component Analysis2002
- Finite Mixture Models2000
- Comparing partitionsJournal of Classification, 1985