Uncertainty in Artificial Intelligence
First Name   Last Name   Password   Forgot Password   Log in!
    Proceedings   Proceeding details   Article details         Authors         Search    
Quantifying Nonlocal Informativeness in High-Dimensional, Loopy Gaussian Graphical Models
Daniel Levine, Jonathan How
Abstract:
We consider the problem of selecting informative observations in Gaussian graphical models con- taining both cycles and nuisances. More specif- ically, we consider the subproblem of quantify- ing conditional mutual information measures that are nonlocal on such graphs. The ability to effi- ciently quantify the information content of obser- vations is crucial for resource-constrained data acquisition (adaptive sampling) and data process- ing (active learning) systems. While closed- form expressions for Gaussian mutual informa- tion exist, standard linear algebraic techniques, with complexity cubic in the network size, are in- tractable for high-dimensional distributions. We investigate the use of embedded trees for com- puting nonlocal pairwise mutual information and demonstrate through numerical simulations that the presented approach achieves a significant re- duction in computational cost over inversion- based methods.
Keywords:
Pages: 487-495
PS Link:
PDF Link: /papers/14/p487-levine.pdf
BibTex:
@INPROCEEDINGS{Levine14,
AUTHOR = "Daniel Levine and Jonathan How",
TITLE = "Quantifying Nonlocal Informativeness in High-Dimensional, Loopy Gaussian Graphical Models",
BOOKTITLE = "Proceedings of the Thirtieth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-14)",
PUBLISHER = "AUAI Press",
ADDRESS = "Corvallis, Oregon",
YEAR = "2014",
PAGES = "487--495"
}


hosted by DSL   •   site info   •   help