Relevance in Probabilistic Models: "Backyards" in a "Small World" (1994)
| Venue: | In Working notes of the AAAI--1994 Fall Symposium Series: Relevance |
| Citations: | 8 - 4 self |
BibTeX
@INPROCEEDINGS{Druzdzel94relevancein,
author = {Marek J. Druzdzel and Henri J. Suermondt},
title = {Relevance in Probabilistic Models: "Backyards" in a "Small World"},
booktitle = {In Working notes of the AAAI--1994 Fall Symposium Series: Relevance},
year = {1994},
pages = {60--63},
publisher = {}
}
Years of Citing Articles
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
Each of the variables in a large probabilistic model may be relevant for some types of reasoning within this model, but rarely will all of them participate in reasoning related to a single query. We review a variety of schemes to identify variables that given certain observations are relevant to a query of interest. Introduction Wer gar zu viel bedenkt wird wenig leisten. (He who considers too much will perform little.) Johann Christoph Friedrich von Schiller "Wilhelm Tell," Act iii, Sc. 1 As outlined carefully by Leonard Savage [ 1972 ] in his influential book on the foundations of Bayesian probability theory and decision theory, probabilistic reasoning is always confined to a well defined set of uncertain variables, which Savage refers to as "small world." Whether the "small world" has been built by a human expert or by a computer, it may include hundreds or thousands of variables, a size that is computationally prohibitive given the complexity of probabilistic inference [ Cooper...
