Computing Upper and Lower Bounds on Likelihoods in Intractable Networks
Tommi Jaakkola, Michael Jordan
We present deterministic techniques for computing upper and lower bounds on marginal probabilities in sigmoid and noisy-OR networks. These techniques become useful when the size of the network (or clique size) precludes exact computations. We illustrate the tightness of the bounds by numerical experiments.
Keywords: Approximate inference, likelihood bounds, Noisy-OR networks, sigmoid networks.
PS Link: ftp://psyche.mit.edu/pub/tommi/jaak-ul-bounds.ps
PDF Link: /papers/96/p340-jaakkola.pdf
AUTHOR = "Tommi Jaakkola
and Michael Jordan",
TITLE = "Computing Upper and Lower Bounds on Likelihoods in Intractable Networks",
BOOKTITLE = "Proceedings of the Twelfth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-96)",
PUBLISHER = "Morgan Kaufmann",
ADDRESS = "San Francisco, CA",
YEAR = "1996",
PAGES = "340--348"