BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:AR Identification of Latent Graphical Models - Mattia Zorzi\, Univ
 ersity of Liege/ University of Cambridge
DTSTART:20140515T130000Z
DTEND:20140515T140000Z
UID:TALK52108@talks.cam.ac.uk
CONTACT:Tim Hughes
DESCRIPTION:Consider an autoregressive gaussian stationary stochastic proc
 ess wherein the manifest variables\, accessible to observations\, are most
 ly related through a limited number of latent variables\, not accessible t
 o observations. It turns out that the inverse of the spectral density of t
 he manifest variables admits a decomposition which is both sparse and low 
 rank. We propose an identification procedure for such processes which expl
 oits the sparse plus low rank decomposition and the efficient convex relax
 ation of such a decomposition.
LOCATION:Cambridge University Engineering Department\, LR6
END:VEVENT
END:VCALENDAR