Fast nonparametric estimation of sparse precision matrices with a priori sparsity structure
Abstract
This work describes a computational approach for rapidly estimating and characterizing the covariance structure of a Gaussian random field from a modest number of i.i.d. replicates by formulating the problem in terms of a sparse inverse Cholesky factor of the data's covariance matrix. Borrowing from the extensive literature of using directed acyclic graphs to impose sparsity on inverse Cholesky factors (for example, Vecchia's approximation or nearest-neighbor Gaussian process models), a probabilistically parsimonious a priori sparsity pattern is imposed, and with careful formulation the estimation problem is given as an extremely well-structured convex problem that can be solved for millions of unknowns in seconds on a single processor core. After describing the problem and specific formulation of the optimization, we demonstrate the estimator's speed and effectiveness at capturing a variety of nonstationary structures in spatial or spatio-temporal data, and in ensemble data assimilation.
Related
- Nonparametric shrinkage estimation of high-dimensional variance component matrices
- Bayesian Nonparametric Factor Analysis via Marginalized Dirichlet Process Column Clustering with Spike-and-Slab Sparsity
- Nonparametric hazard rate estimation with associated kernels and minimax bandwidth choice
- Exponent Profiles of Oscillatory Matrices under Matrix and Inverse Sparsity
- Oracle high-dimensional M-estimation using smooth reparameterization for sparsity
BibTeX
@misc{derrickson2026fast,
title = {{Fast nonparametric estimation of sparse precision matrices with a priori sparsity structure}},
author = {Caleb Derrickson and Alexis Montoison and Vishwas Rao and Christopher J. Geoga and Mihai Anitescu},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.03462},
eprint = {2611.03462},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.03462},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Fast nonparametric estimation of sparse precision matrices with a priori sparsity structure
AU - Caleb Derrickson
AU - Alexis Montoison
AU - Vishwas Rao
AU - Christopher J. Geoga
AU - Mihai Anitescu
PY - 2026
DA - 2026/11/02
PB - arXiv
DO - 10.48550/arxiv.2611.03462
UR - https://arxiv.org/abs/2611.03462
AB - This work describes a computational approach for rapidly estimating and characterizing the covariance structure of a Gaussian random field from a modest number of i.i.d. replicates by formulating the problem in terms of a sparse inverse Cholesky factor of the data's covariance matrix. Borrowing from the extensive literature of using directed acyclic graphs to impose sparsity on inverse Cholesky factors (for example, Vecchia's approximation or nearest-neighbor Gaussian process models), a probabilistically parsimonious a priori sparsity pattern is imposed, and with careful formulation the estimation problem is given as an extremely well-structured convex problem that can be solved for millions of unknowns in seconds on a single processor core. After describing the problem and specific formulation of the optimization, we demonstrate the estimator's speed and effectiveness at capturing a variety of nonstationary structures in spatial or spatio-temporal data, and in ensemble data assimilation.
ER -