Statistics – Machine Learning
Scientific paper
2011-11-23
Statistics
Machine Learning
Scientific paper
The graphical lasso [Banerjee et al., 2008, Friedman et al., 2007b] is a popular approach for learning the structure in an undirected Gaussian graphical model, using $\ell_1$ regularization to control the number of zeros in the precision matrix ?${\B\Theta}={\B\Sigma}^{-1}$. The R package glasso is popular, fast, and allows one to efficiently build a path of models for different values of the tuning parameter. Convergence of GLASSO can be tricky; the converged precision matrix might not be the inverse of the estimated covariance, and occasionally it fails to converge with warm starts. In this paper we explain this behavior, and propose new algorithms that appear to outperform GLASSO. We show that in fact glasso is solving the dual of the graphical lasso penalized likelihood, by block coordinate descent. In this dual, the target of estimation is $\B\Sigma$, the covariance matrix, rather than the precision matrix $\B\Theta$. We propose similar primal algorithms P-GLASSO and DP-GLASSO, that also operate by block-coordinate descent, where $\B\Theta$ is the optimization target. We study all of these algorithms, and in particular different approaches to solving their coordinate subproblems. We conclude that DP-GLASSO is superior from several points of view.
Hastie Trevor
Mazumder Rahul
No associations
LandOfFree
The Graphical Lasso: New Insights and Alternatives does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with The Graphical Lasso: New Insights and Alternatives, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and The Graphical Lasso: New Insights and Alternatives will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-377204