You are here: Home / Publications / The joint graphical lasso for inverse covariance estimation across multiple classes.

The joint graphical lasso for inverse covariance estimation across multiple classes.

24817823

J R Stat Soc Series B Stat Methodol. 2014 Mar 76 (2).

joint graphical lasso, which borrows strength across the classes in order to estimate multiple graphical models that share certain characteristics, such as the locations or weights of nonzero edges. Our approach is based upon maximizing a penalized log likelihood. We employ generalized fused lasso or group lasso penalties, and implement a fast ADMM algorithm to solve the corresponding convex optimization problems. The performance of the proposed method is illustrated through simulated and real data examples.

This icon signifies that something is happening and we kindly ask you to please wait