Sparse inverse covariance estimation with the graphical lasso - PubMed (original) (raw)

Sparse inverse covariance estimation with the graphical lasso

Jerome Friedman et al. Biostatistics. 2008 Jul.

Abstract

We consider the problem of estimating sparse graphs by a lasso penalty applied to the inverse covariance matrix. Using a coordinate descent procedure for the lasso, we develop a simple algorithm--the graphical lasso--that is remarkably fast: It solves a 1000-node problem ( approximately 500,000 parameters) in at most a minute and is 30-4000 times faster than competing methods. It also provides a conceptual link between the exact problem and the approximation suggested by Meinshausen and Bühlmann (2006). We illustrate the method on some cell-signaling data from proteomics.

PubMed Disclaimer

Figures

Fig. 1.

Number of CPU seconds required for the graphical lasso procedure.

Fig. 2.

Directed acylic graph from cell-signaling data, from Sachs and others (2003).

Fig. 3.

Cell-signaling data: Undirected graphs from graphical lasso with different values of the penalty parameter ρ.

Fig. 4.

Cell-signaling data: Profile of coefficients as the total_L_1norm of the coefficient vector increases, that is as_ρ_decreases. Profiles for the largest coefficients are labeled with the corresponding pair of proteins.

Fig. 5.

Cell-signaling data. Left panel shows 10-fold cross-validation using both regression and likelihood approaches (details in text). Right panel compares the regression sum of squares of the exact graphical lasso approach to the Meinhausen–Buhlmann approximation.

Cited by

Inference on the Eigenvalues of the Normalized Precision Matrix.
Duttweiler L, Almudevar A. Duttweiler L, et al. Linear Algebra Appl. 2024 Dec 15;703:78-108. doi: 10.1016/j.laa.2024.09.002. Epub 2024 Sep 10. Linear Algebra Appl. 2024. PMID: 39372671
Using network analysis to provide evidence for brain health as a unified construct relevant to aging with HIV.
Matout M, Brouillette MJ, Fellows LK, Mayo NE. Matout M, et al. Soc Psychiatry Psychiatr Epidemiol. 2024 Oct 5. doi: 10.1007/s00127-024-02778-z. Online ahead of print. Soc Psychiatry Psychiatr Epidemiol. 2024. PMID: 39368026
Machine learning and Bayesian network analyses identifies associations with insomnia in a national sample of 31,285 treatment-seeking college students.
Calderon A, Baik SY, Ng MHS, Fitzsimmons-Craft EE, Eisenberg D, Wilfley DE, Taylor CB, Newman MG. Calderon A, et al. BMC Psychiatry. 2024 Oct 5;24(1):656. doi: 10.1186/s12888-024-06074-7. BMC Psychiatry. 2024. PMID: 39367432 Free PMC article.
Associations between Biological Maturation, Chronological Age, Body Mass Index, Sex, and Motor Competence in Prepubertal Children: A Network Analysis.
Salami S, Bandeira PFR, Mashhadi MR, Shamsipour Dehkordi P, Luz LGO, Martins CML, Hardy LL, Duncan MJ. Salami S, et al. Children (Basel). 2024 Sep 20;11(9):1143. doi: 10.3390/children11091143. Children (Basel). 2024. PMID: 39334675 Free PMC article.
Is Seeing Believing? A Practitioner's Perspective on High-Dimensional Statistical Inference in Cancer Genomics Studies.
Fan K, Subedi S, Yang G, Lu X, Ren J, Wu C. Fan K, et al. Entropy (Basel). 2024 Sep 16;26(9):794. doi: 10.3390/e26090794. Entropy (Basel). 2024. PMID: 39330127 Free PMC article. Review.

References

1. Banerjee O, Ghaoui LE, d'Aspremont A. Model selection through sparse maximum likelihood estimation. Journal of Machine Learning Research. 2007 101 (to appear)
1. Boyd S, Vandenberghe L. Convex Optimization. Cambridge: Cambridge University Press; 2004.
1. Dahl J, Vandenberghe L, Roychowdhury V. Covariance selection for non-chordal graphs via chordal embedding. Optimization Methods and Software (to appear) 2007
1. Friedman J, Hastie T, Hoefling H, Tibshirani R. Pathwise coordinate optimization. Annals of Applied Statistics. 2007;2
1. Meinshausen N, Bühlmann P. High dimensional graphs and variable selection with the lasso. Annals of Statistics. 2006;34:1436–1462.

Sparse inverse covariance estimation with the graphical lasso - PubMed (original) (raw)