Module-based regularization improves Gaussian graphical models when observing noisy data

2023/03/29 by Magnus Neuman, Joaquín Calatayud, Neuman, Magnus +5 · 1 citation
Biochemistry, Genetics and Molecular Biology · Computer Science · #Bayesian Modeling and Causal Inference #Bioinformatics and Genomic Networks #Data Analysis #FOS: Computer and information sciences #FOS: Physical sciences #Gene expression and cancer classification #Machine Learning (cs.LG) #Statistics and Probability (physics.data-an)

paper · pdf · doi:10.48550/arxiv.2303.16796

openalex publication_date 2023/03/29 · openalex created_date 2023/04/05 · openalex updated_date 2026/07/28

Abstract

Inferring relations from correlational data allows researchers across the sciences to uncover complex connections between variables for insights into the underlying mechanisms. The researchers often represent inferred relations using Gaussian graphical models, requiring regularization to sparsify the models. Acknowledging that the modular structure of the inferred network is often studied, we suggest module-based regularization to balance under- and overfitting. Compared with the graphical lasso, a standard approach using the Gaussian log-likelihood for estimating the regularization strength, this approach better recovers and infers modular structure in noisy synthetic and real data. The module-based regularization technique improves the usefulness of Gaussian graphical models in the many applications where they are employed.

Cited by

• Reliable data clustering with Bayesian community detection

Related

• Selection of the Regularization Parameter in Graphical Models using Network Characteristics
• Brief Report on Estimating Regularized Gaussian Networks from Continuous and Ordinal Data
• Structured Regularization for conditional Gaussian Graphical Models
• Regularization for supervised learning via the "hubNet" procedure
• The Cluster Graphical Lasso for improved estimation of Gaussian graphical models