{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

NIPS2009_0023_slide - Results an l 1-norm penalty for local...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
Sparse and Locally Constant Gaussian Graphical Models M3 Jean Honorio, Luis Ortiz, Dimitris Samaras, Nikos Paragios, Rita Goldstein Motivation: learning GGMs on spatial datasets, e.g.: silhouettes, motion trajectories, 2D and 3D images, with spatial coherence of the dependence/independence relationships (local constancy)
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Results: an l 1-norm penalty for local constancy in a strictly convex maximum likelihood estimation, and an efficient algorithm that decomposes the original optimization problem into a sequence of non-smooth piecewise quadratic problems with closed form solutions walking sequence cardiac MRI brain fMRI...
View Full Document

{[ snackBarMessage ]}