Efficient Search-Based Inference for noisy-OR Belief Networks: TopEpsilon

Inference algorithms for arbitrary belief networks are impractical for large, complex belief networks. Inference algorithms for specialized classes of belief networks have been shown to be more efficient. In this paper, we present a searchbased algorithm for approximate inference on arbitrary, noisy-OR belief networks, generalizing earlier work on search-based inference for twolevel, noisy-OR belief networks. Initial experimental results appear promising.

• Type: Conference
• Year: 1996
• Where: UAI
• Authors: Kurt Huang, Max Henrion
• Keywords: Belief Networks, Inference Algorithms, Noisy-OR Belief Networks, UAI 1996, UAI 2008