ContextSpecific Independence in Bayesian Networks
Craig Boutilier, Nir Friedman, Moises Goldszmidt, Daphne Koller
Abstract:
Bayesian networks provide a language for qualitatively representing the conditional independence properties of a distribution. This allows a natural and compact representation of the distribution, eases knowledge acquisition, and supports effective inference algorithms. It is wellknown, however, that there are certain independencies that we cannot capture qualitatively within the Bayesian network structure: independencies that hold only in certain contexts, i.e., given a specific assignment of values to certain variables. In this paper, we propose a formal notion of contextspecific independence (CSI), based on regularities in the conditional probability tables (CPTs) at a node. We present a technique, analogous to (and based on) dseparation, for determining when such independence holds in a given network. We then focus on a particular qualitative representation schemetreestructured CPTsfor capturing CSI. We suggest ways in which this representation can be used to support effective inference algorithms. In particular, we present a structural decomposition of the resulting network which can improve the performance of clustering algorithms, and an alternative algorithm based on cutset conditioning.
Keywords: Conditional independence, decision trees, probabilistic inference,
cutset conditioni
Pages: 115123
PS Link: http://robotics.stanford.edu/~koller/papers/uai96.ps
PDF Link: /papers/96/p115boutilier.pdf
BibTex:
@INPROCEEDINGS{Boutilier96,
AUTHOR = "Craig Boutilier
and Nir Friedman and Moises Goldszmidt and Daphne Koller",
TITLE = "ContextSpecific Independence in Bayesian Networks",
BOOKTITLE = "Proceedings of the Twelfth Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI96)",
PUBLISHER = "Morgan Kaufmann",
ADDRESS = "San Francisco, CA",
YEAR = "1996",
PAGES = "115123"
}

