University of Cambridge > > CUED Control Group Seminars > AR Identification of Latent Graphical Models

AR Identification of Latent Graphical Models

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Tim Hughes.

Consider an autoregressive gaussian stationary stochastic process wherein the manifest variables, accessible to observations, are mostly related through a limited number of latent variables, not accessible to observations. It turns out that the inverse of the spectral density of the manifest variables admits a decomposition which is both sparse and low rank. We propose an identification procedure for such processes which exploits the sparse plus low rank decomposition and the efficient convex relaxation of such a decomposition.

This talk is part of the CUED Control Group Seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2022, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity