An empirical analysis of likelihood-weighting simulation on a large, multiply-connected belief network
From Semantic Portal Wiki
{{#vardefine:category|Publication}}{{#vardefine:templatename|i.publication}}{{#vardefine:package|smwbp_instance_templates}}
| Edit |
Reference: {{#vardefine:pagename|an empirical analysis of likelihood-weighting simulation on a large, multiply-connected belief network }}
- [[]]
bibtex
{{#vardefine:pagename|An empirical analysis of likelihood-weighting simulation on a large, multiply-connected belief network }}{{#vardefine:key| }}
abstract: We analyzed the convergence properties of likelihood-weighting algorithms on a two-level, multiply connected, belief-network representation of the QMR knowledge base of internal medicine. Specifically, on two difficult diagnostic cases, we examined the effects of Markov blanket scoring, importance sampling, and self-importance sampling, demonstrating that the simulation on our model requires the Markov blanket scoring and self-importance sampling to converge well.
download:
- paper:
- slides:
| Abstract | We analyzed the convergence properties of … We analyzed the convergence properties of likelihood-weighting algorithms on a two-level, multiply connected, belief-network representation of the QMR knowledge base of internal medicine. Specifically, on two difficult diagnostic cases, we examined the effects of Markov blanket scoring, importance sampling, and self-importance sampling, demonstrating that the simulation on our model requires the Markov blanket scoring and self-importance sampling to converge well. self-importance sampling to converge well. |
| Author | Michael Shwe +, and Gregory F. Cooper + |
| Bibtype | techreport + |
| Institution | Knowledge Systems, AI Laboratory + |
| Key | KSL-90-23 + |
| Number | KSL-90-23 + |
| Tag | Computer science + |
| Title | An Empirical Analysis of Likelihood-Weighting Simulation on a Large, Multiply-Connected Belief Network + |
| Tr id | KSL-90-23 + |
| Year | 1991 + |
