Uncertainty in Artificial Intelligence
First Name   Last Name   Password   Forgot Password   Log in!
    Proceedings         Authors   Author's Info   Article details         Search    
Importance Sampling via Variational Optimization
Ydo Wexler, Dan Geiger
Abstract:
Computing the exact likelihood of data in large Bayesian networks consisting of thousands of vertices is often a difficult task. When these models contain many deterministic conditional probability tables and when the observed values are extremely unlikely even alternative algorithms such as variational methods and stochastic sampling often perform poorly. We present a new importance sampling algorithm for Bayesian networks which is based on variational techniques. We use the updates of the importance function to predict whether the stochastic sampling converged above or below the true likelihood, and change the proposal distribution accordingly. The validity of the method and its contribution to convergence is demonstrated on hard networks of large genetic linkage analysis tasks.
Keywords:
Pages: 426-433
PS Link:
PDF Link: /papers/07/p426-wexler.pdf
BibTex:
@INPROCEEDINGS{Wexler07,
AUTHOR = "Ydo Wexler and Dan Geiger",
TITLE = "Importance Sampling via Variational Optimization",
BOOKTITLE = "Proceedings of the Twenty-Third Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-07)",
PUBLISHER = "AUAI Press",
ADDRESS = "Corvallis, Oregon",
YEAR = "2007",
PAGES = "426--433"
}


hosted by DSL   •   site info   •   help