Filters








22 Hits in 5.9 sec

A Comparison of Lauritzen-Spiegelhalter, Hugin, and Shenoy-Shafer Architectures for Computing Marginals of Probability Distributions [article]

Vasilica Lepar, Prakash P. Shenoy
2013 arXiv   pre-print
In this paper, we compare three architectures - Lauritzen-Spiegelhalter, Hugin, and Shenoy-Shafer - from the perspective of graphical structure for message propagation, message-passing scheme, computational  ...  In the last decade, several architectures have been proposed for exact computation of marginals using local computation.  ...  Acknowledgments The authors are grateful for support and encouragement from Professor Jiirg Kohlas. We are especially grateful to Robert SHirk and Norbert Lehmann for their help on this project.  ... 
arXiv:1301.7394v1 fatcat:lwnw3yi3hvaaldlpjltabaijsm

Some improvements to the Shenoy-Shafer and Hugin architectures for computing marginals

Tuija Schmidt, Prakash P. Shenoy
1998 Artificial Intelligence  
The main aim of this paper is to describe two modifications to the Shenoy-Shafer architecture with the goal of making it computationally more efficient in computing marginals of the joint valuation.  ...  We also describe a modification to the Hugin architecture.  ...  Acknowledgements The authors are grateful to Vasilica Lepar for comments and discussion of the paper.  ... 
doi:10.1016/s0004-3702(98)00047-2 fatcat:ikz3p3hmxna3pkbsxal7swlt7i

Computation in Valuation Algebras [chapter]

Jürg Kohlas, Prakash P. Shenoy
2000 Handbook of Defeasible Reasoning and Uncertainty Management Systems  
The additional notion of continuers is introduced and, based on it, two more computational architectures, the Lauritzen-Spiegelhalter and the HUGIN architecture, are presented.  ...  Based on this algebraic structure, different inference mechanisms that use local computations are described. These include the fusion algorithm and, derived from it, the Shenoy-Shafer architecture.  ...  Sections 4 and 5 present different computational architectures, namely Shafer-Shenoy architecture, Lauritzen-Spiegelhalter architecture, and HUGIN architecture.  ... 
doi:10.1007/978-94-017-1737-3_2 fatcat:ajqhkygcizfobbqnuwp7z7kdnq

Nested Junction Trees [article]

Uffe Kjærulff
2013 arXiv   pre-print
The efficiency of inference in both the Hugin and, most notably, the Shafer-Shenoy architectures can be improved by exploiting the independence relations induced by the incoming messages of a clique.  ...  That is, the message to be sent from a clique can be computed via a factorization of the clique potential in the form of a junction tree.  ...  Lauritzen for suggesting the cost propagation scheme, Claus S. Jensen for provid ing the Link and Pignet networks, David Beckerman for providing the Pathfinder network, Kristian G.  ... 
arXiv:1302.1553v1 fatcat:jkxgqe75obaqniqpvxz3bs5hd4

FAULT-TOLERANT MULTI-AGENT EXACT BELIEF PROPAGATION

Xiangdong An, Nick Cercone
2009 Computational intelligence  
In earlier work, all belief updating methods on a hypertree are made of two rounds of propagation, each of which is implemented as a recursive process.  ...  In this paper, we present a fault-tolerant belief updating method for multiagent probabilistic inference.  ...  They are Lauritzen-Spiegelhalter architecture, Hugin architecture, Shenoy-Shafer architecture, and Lazy architecture.  ... 
doi:10.1111/j.1467-8640.2008.01328.x fatcat:asjo7g67rzh3th2z7z25vmf3xq

Belief update in CLG Bayesian networks with lazy propagation

A.L. Madsen
2008 International Journal of Approximate Reasoning  
Jensen, Stable local computation with mixed Gaussian distributions, Statistics and Computing 11 (2) (2001) 191-203] and Cowell [R.G.  ...  The proposed architecture is an extension of lazy propagation using operations of Lauritzen and Jensen [S.L. Lauritzen, F.  ...  Acknowledgement This paper is an extended and revised version of [15] .  ... 
doi:10.1016/j.ijar.2008.05.001 fatcat:cqhkcskbxnbt7nhe73327xqtyq

Lazy propagation: A junction tree inference algorithm based on lazy evaluation

Anders L. Madsen, Finn V. Jensen
1999 Artificial Intelligence  
We compare the time and space performance of the proposed architecture with non-optimized implementations of the HUGIN and Shafer-Shenoy inference architectures.  ...  The tasks we consider include cautious propagation of evidence, determining a most probable configuration, and fast retraction of evidence a long with a number of other tasks.  ...  Madsen was a PhD-student at Department of Computer Science, Aalborg University, Denmark. This research was partly supported by the Danish Natural Science Research Council grant 9601649.  ... 
doi:10.1016/s0004-3702(99)00062-4 fatcat:24ua53bsvbfvrph4wqpnvz3xpq

Inference in hybrid Bayesian networks using dynamic discretization

Martin Neil, Manesh Tailor, David Marquez
2007 Statistics and computing  
Our approach offers a significant extension to Bayesian Network theory and practice by offering a flexible way of modelling continuous nodes in BNs conditioned on complex configurations of evidence and  ...  In particular, we show how the rapid convergence of the algorithm towards zones of high probability density, make robust inference analysis possible even in situations where, due to the lack of information  ...  to the choice of the Hugin architecture as a platform to compute the marginal distributions.  ... 
doi:10.1007/s11222-007-9018-y fatcat:f744kwtpnjaj5ibaftgbmfe6aa

Lazy Probability Propagation on Gaussian Bayesian Networks

Hua Mu, Meiping Wu, Hongxu Ma, Tim Bailey
2010 2010 22nd IEEE International Conference on Tools with Artificial Intelligence  
Novel lazy Lauritzen-Spiegelhalter (LS), lazy Hugin and lazy Shafer-Shenoy (SS) algorithms are devised for Gaussian Bayesian networks (BNs).  ...  The moments form parametrization of Gaussian distributions allows the deterministic relationships between variables.  ...  We plan to evaluate empirically the proposed algorithms and compare them with some existent schemes, like that in [13] , through a wide range of Gaussian BNs in the future.  ... 
doi:10.1109/ictai.2010.51 dblp:conf/ictai/MuWMB10 fatcat:li4g2nsagjb5ldaruppvfdud5y

A Review of Inference Algorithms for Hybrid Bayesian Networks

Antonio Salmerón, Rafael Rumí, Helge Langseth, Thomas D. Nielsen, Anders L. Madsen
2018 The Journal of Artificial Intelligence Research  
In this paper we provide an overview of the main trends and principled approaches for performing inference in hybrid Bayesian networks.  ...  However, this extra feature also comes at a cost: inference in these types of models is computationally more challenging and the underlying models and updating procedures may not even support closed-form  ...  This research has been partly funded by the Spanish Ministry of Economy and Competitiveness, through project TIN2016-77902-C3-3-P and by ERDF funds.  ... 
doi:10.1613/jair.1.11228 fatcat:vhmuf44ftbg73mjygiyrwrdw44

An Evolutionary Algorithm for Bayesian Network Triangulation [chapter]

Tomasz Łukaszewski
2003 Operations Research Proceedings 2002  
A local search heuristic based on the idea of evolutionary algorithms is presented. The results obtained using existing and proposed approaches are compared on a basis of a computational experiment.  ...  The problem of triangulation (decomposition) of Bayesian networks is considered. Triangularity of a Bayesian network is required in a general evidence propagation scheme on this network.  ...  The best known algorithms are Lauritzen-Spiegelhalter, Hugin, Shafer-Shenoy [5, 1, 7] . They differ mainly in the construction of messages.  ... 
doi:10.1007/978-3-642-55537-4_59 dblp:conf/or/Lukaszewski02 fatcat:4tecx4eix5a75p7estzukd336y

Efficient computation for the noisy MAX

Francisco J. Díez, Severino F. Galán
2003 International Journal of Intelligent Systems  
Any other marginal or conditional probability can be obtained from it. In particular, it is possible to obtain the a posteriori probability of any variable given a certain evidence: P (v i |e).  ...  However, the straightforward method that computes the conditional probabilities  ...  We assigned a noisy MAX distribution to each family (in fact, it was a noisy OR, since all the variables were binary) and propagated evidence in a Shafer-Shenoy architecture, first with expanded CPT's  ... 
doi:10.1002/int.10080 fatcat:3uibgmb3q5dbvaxovg4rjpclqu

Discrete Bayesian Networks: The Exact Posterior Marginal Distributions [article]

Do Le Paul Minh
2014 arXiv   pre-print
In a Bayesian network, we wish to evaluate the marginal probability of a query variable, which may be conditioned on the observed values of some evidence variables.  ...  the size of network, and linear with the number of its evidence and query variables.  ...  The Hugin architecture has the same concern as the Lauritzen-Spiegelhalter architecture, namely, the size of the clique in a junction tree.  ... 
arXiv:1411.6300v1 fatcat:jfe2o7calffbhk3jeoflvla4hq

Computational algorithm for dynamic hybrid Bayesian network in on-line system health management applications

Chonlagarn Iamsumang, Ali Mosleh, Mohammad Modarres
2014 2014 International Conference on Prognostics and Health Management  
The scope of this research includes a new modeling approach, computation algorithm, and an example application for on-line SHM.  ...  Markov Chain Monte Carlo (MCMC) inference is optimized using a pre-computation strategy and dynamic programming for on-line monitoring of system health.  ...  There are not enough words to describe how grateful and thankful I am for everything they have done for me during these years.  ... 
doi:10.1109/icphm.2014.7036384 fatcat:r3o5labydzdwrii4gmjognisc4

Generalized constraint-based inference

Le Chang
2005
Solving them efficiently is important for both research and practical applications.  ...  Second, the proposed semiring-based unified framework is also a single formal algorithmic framework that provides a broader coverage of both exact and approximate inference algorithms, including variable  ...  Acknowledgements I would like to thank my supervisor, Professor Alan Mackworth, for his guidance and support in this work. I would not be here without his inspiration and encouragement.  ... 
doi:10.14288/1.0051112 fatcat:dux2g44lijhihdwbgh6v5ejdvu
« Previous Showing results 1 — 15 out of 22 results