Learning the structure of dynamic probabilistic networks (1998)
Cached
Download Links
| Citations: | 161 - 13 self |
BibTeX
@INPROCEEDINGS{Friedman98learningthe,
author = {Nir Friedman and Kevin Murphy and Stuart Russell},
title = {Learning the structure of dynamic probabilistic networks},
booktitle = {},
year = {1998},
pages = {139--147},
publisher = {Morgan Kaufmann}
}
Years of Citing Articles
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
Dynamic probabilistic networks are a compact representation of complex stochastic processes. In this paper we examine how to learn the structure of a DPN from data. We extend structure scoring rules for standard probabilistic networks to the dynamic case, and show how to search for structure when some of the variables are hidden. Finally, we examine two applications where such a technology might be useful: predicting and classifying dynamic behaviors, and learning causal orderings in biological processes. We provide empirical results that demonstrate the applicability of our methods in both domains. 1
