Partition-based Anytime Approximation for Belief

@inproceedings{Mateescu2001PartitionbasedAA,
  title={Partition-based Anytime Approximation for Belief},
  author={UpdatingRobert Mateescu},
  year={2001}
}
  • UpdatingRobert Mateescu
  • Published 2001
The paper presents a parameterized approximation scheme for probabilistic inference. The scheme, called Mini-Clustering (MC) extends the partition-based approximation offered by mini-bucket elimination, to tree de-compositions. The beneet of this extension is that all single variable beliefs are computed (approximately) at once, using a two-phase message-passing process along the cluster tree. The resulting approximation scheme is governed by a controlling parameter, "z" that allows adjustable… CONTINUE READING

References

Publications referenced by this paper.
Showing 1-10 of 12 references

Bucket Elimination: A Unifying Framework for Reasoning

Artif. Intell. • 1999
View 4 Excerpts
Highly Influenced

probabilistic reasoning in intelligent systems: networks of plausible inference san mateo

Morgan Kaufmann series in representation and reasoning • 1988
View 3 Excerpts
Highly Influenced

Local computation with probabilities on graphical structures and their applicationto expert systems

S. L. Lauritzen andD. J. Spiegelhalter
Journal of the Royal StatisticalSociety , Series B • 2001

A general algorithm for approximate inference and its applciation to hybrid bayes nets

K. Kask J Larrosa
InProceedings of the 14 th Conference on Uncertaintyin Arti cial Intelligence • 1998

Treeclustering for constraint networks

J. Pearl.
1989

Similar Papers

Loading similar papers…