Advanced search
1 file | 128.92 KB

Belief propagation in imprecise Markov trees

Gert De Cooman (UGent) , Filip Hermans (UGent) and Erik Quaeghebeur (UGent)
Author
Organization
Abstract
We replace strong independence in credal networks with the weaker notion of epistemic irrelevance. Focusing on directed trees, we show how to combine the local credal sets in the networks into an overall joint model, and use this to construct and justify an exact message-passing algorithm that computes updated beliefs for a variable in the network. The algorithm, which is essentially linear in the number of nodes, is formulated entirely in terms of coherent lower previsions. We supply examples of the algorithm's operation, and report an application to on-line character recognition that illustrates the advantages of the model for prediction.

Downloads

  • FH-2009-DYSCO-poster.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 128.92 KB

Citation

Please use this url to cite or link to this publication:

Chicago
De Cooman, Gert, Filip Hermans, and Erik Quaeghebeur. 2009. “Belief Propagation in Imprecise Markov Trees.” In Interuniversity Attraction Pole IAP VI/4 DYSCO Study Day, Posters.
APA
De Cooman, Gert, Hermans, F., & Quaeghebeur, E. (2009). Belief propagation in imprecise Markov trees. Interuniversity Attraction Pole IAP VI/4 DYSCO Study Day, Posters. Presented at the Interuniversity Attraction Pole IAP VI/4 DYSCO Study Day.
Vancouver
1.
De Cooman G, Hermans F, Quaeghebeur E. Belief propagation in imprecise Markov trees. Interuniversity Attraction Pole IAP VI/4 DYSCO Study Day, Posters. 2009.
MLA
De Cooman, Gert, Filip Hermans, and Erik Quaeghebeur. “Belief Propagation in Imprecise Markov Trees.” Interuniversity Attraction Pole IAP VI/4 DYSCO Study Day, Posters. 2009. Print.
@inproceedings{1974578,
  abstract     = {We replace strong independence in credal networks with the weaker notion of epistemic irrelevance. Focusing on directed trees, we show how to combine the local credal sets in the networks into an overall joint model, and use this to construct and justify an exact message-passing algorithm that computes updated beliefs for a variable in the network. The algorithm, which is essentially linear in the number of nodes, is formulated entirely in terms of coherent lower previsions. We supply examples of the algorithm's operation, and report an application to on-line character recognition that illustrates the advantages of the model for prediction.},
  author       = {De Cooman, Gert and Hermans, Filip and Quaeghebeur, Erik},
  booktitle    = {Interuniversity Attraction Pole IAP VI/4 DYSCO Study Day, Posters},
  language     = {eng},
  location     = {Mons, Belgium},
  title        = {Belief propagation in imprecise Markov trees},
  year         = {2009},
}