|
KU ScholarWorks >
School of Business >
School of Business Working Papers >
Please use this identifier to cite or link to this item:
http://hdl.handle.net/1808/467
|
|
View usage statistics
|
| Title: | Inference in Hybrid Bayesian Networks with Mixtures of Truncated Exponentials |
| Authors: | Cobb, Barry R.
Shenoy, Prakash P. |
| Keywords: | Hybrid Bayesian networks
mixtures of truncated exponentials
Shenoy-Shafer architecture
conditional linear Gaussian models |
| Issue Date: | Jun-2005 |
| Publisher: | University of Kansas School of Business |
| Extent: | 443794 bytes |
| Type: | Working Paper |
| Citation: | Cobb, B. R. and P. P. Shenoy (2005), "Inference in Hybrid Bayesian Networks with Mixtures of Truncated Exponentials," University of Kansas School of Business Working Paper No. 294, July 2003, Revised June 2005, Lawrence, KS. |
| Series/Report no.: | School of Business Working Paper;294 |
| Abstract: | Mixtures of truncated exponentials (MTE) potentials are an alternative to discretization for solving hybrid Bayesian networks. Any probability density function (PDF) can be approximated with an MTE potential, which can always be marginalized in closed form. This allows propagation to be done exactly using the Shenoy-Shafer architecture for computing marginals, with no restrictions on the construction of a join tree. This paper presents MTE potentials that approximate an arbitrary normal PDF with any mean and a positive variance. The properties of these MTE potentials are presented, along with examples that demonstrate their use in solving hybrid Bayesian networks. Assuming that the joint density exists, MTE potentials can be used for inference in hybrid Bayesian networks that do not fit the restrictive assumptions of the conditional linear Gaussian (CLG) model, such as networks containing discrete nodes with continuous parents. |
| Description: | Has been accepted for publication in the International Journal of Approximate Reasoning, Elsevier Science Publishing Co., Inc. |
| URI: | http://hdl.handle.net/1808/467 |
| Appears in Collections: | School of Business Working Papers
|
Items in KU ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
|
top