Bayesian Graphical Models under Positivity Constraints: A Scalable Generalized Likelihood Approach
Abstract
We develop a computationally scalable Bayesian framework for precision matrix estimation in Gaussian graphical models under total positivity constraints. To overcome the high computational cost of the Gaussian likelihood, we adopt a generalized Bayesian approach based on the D-trace loss, which eliminates the log-determinant term and enables efficient optimization while allowing relaxation of positive definiteness during sampling. Sparsity is induced via spike-and-slab priors, and the resulting generalized posterior is shown to be proper under mild conditions. Our primary contribution is a suite of efficient posterior sampling algorithms tailored to high-dimensional settings. Starting from a component-wise Gibbs sampler, we introduce a novel data augmentation scheme that induces conditional independence among precision matrix entries, enabling joint updates. By exploiting the Gram structure of the sample covariance matrix, we further develop a fast matrix-normal sampler that significantly reduces per-iteration complexity in high-dimensional settings. An interweaving strategy combines augmented and direct updates to improve mixing without sacrificing scalability. Experiments on synthetic and financial data demonstrate substantial computational gains over existing methods, while maintaining competitive estimation accuracy and improved recovery of structured dependencies.
Related
- Penalized Likelihood Inference in Highly Multivariate Sparse Spectral Graphical Gaussian Processes
- A computational approach to maximum likelihood thresholds for colored Gaussian graphical models
- Comment on Vogels et al.'s (2024) "Bayesian Structure Learning in Undirected Gaussian Graphical Models: Literature Review with Empirical Comparison": An updated performance check using BGGM
- Analysis and Reporting Guidelines for Bayesian Graphical Modeling of Cross-Sectional Data: A Tutorial in JASP
- Graphical Discreteness, Coxeter Doublings and Generalized Polygons
BibTeX
@misc{raha2026bayesian,
title = {{Bayesian Graphical Models under Positivity Constraints: A Scalable Generalized Likelihood Approach}},
author = {Swarnali Raha and Partha Sarkar and Sirani M. Perera and Kshitij Khare},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2607.26469},
eprint = {2607.26469},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.26469},
}
RIS
TY - UNPB
TI - Bayesian Graphical Models under Positivity Constraints: A Scalable Generalized Likelihood Approach
AU - Swarnali Raha
AU - Partha Sarkar
AU - Sirani M. Perera
AU - Kshitij Khare
PY - 2026
DA - 2026/10/22
PB - arXiv
DO - 10.48550/arxiv.2607.26469
UR - https://arxiv.org/abs/2607.26469
AB - We develop a computationally scalable Bayesian framework for precision matrix estimation in Gaussian graphical models under total positivity constraints. To overcome the high computational cost of the Gaussian likelihood, we adopt a generalized Bayesian approach based on the D-trace loss, which eliminates the log-determinant term and enables efficient optimization while allowing relaxation of positive definiteness during sampling. Sparsity is induced via spike-and-slab priors, and the resulting generalized posterior is shown to be proper under mild conditions. Our primary contribution is a suite of efficient posterior sampling algorithms tailored to high-dimensional settings. Starting from a component-wise Gibbs sampler, we introduce a novel data augmentation scheme that induces conditional independence among precision matrix entries, enabling joint updates. By exploiting the Gram structure of the sample covariance matrix, we further develop a fast matrix-normal sampler that significantly reduces per-iteration complexity in high-dimensional settings. An interweaving strategy combines augmented and direct updates to improve mixing without sacrificing scalability. Experiments on synthetic and financial data demonstrate substantial computational gains over existing methods, while maintaining competitive estimation accuracy and improved recovery of structured dependencies.
ER -