Confusion related to dual problem formulation in sparse inverse covariance matrix estimation

I was reading this paper where they are trying to estimate the inverse covariance matrix of the gaussian. What they are trying to maximize the gaussian log likelihood. The primal problem is maximize the likelihood wrt the inverse covariance matrix.

I didn't get how the dual was formed. Can anyone provide some hints or guidance?

enter image description here

The author also has a talk available at . However, she doesn't describe how she got this dual as well.

asked May 25 '13 at 00:13

Jason Tyler
16●13●13●15

edited May 25 '13 at 00:18

Subscription:

Once you sign in you will be able to subscribe for any updates here

Tags:

optimization ×136
gaussian ×17
likelihood ×15
precision ×11
covariance ×7
duality ×5

Asked: May 25 '13 at 00:13

Seen: 710 times

Last updated: May 25 '13 at 00:18

Confusion related to dual problem formulation in sparse inverse covariance matrix estimation

Subscription:

Related questions