In this paper we demonstrate how Gr�bner bases and other algebraic
techniques can be used to explore the geometry of the probability
space of Bayesian networks with hidden variables. These techniques
employ a parametrisation of Bayesian network by moments rather than
conditional probabilities. We show that whilst Gr�bner bases help
to explain the local geometry of these spaces a complimentary analysis,
modelling the positivity of probabilities, enhances and completes
the geometrical picture. We report some recent geometrical results
in this area and discuss a possible general methodology for the analyses
of such problems.
%0 Journal Article
%1 Smith2003b
%A Smith, J. Q.
%A Croft, J.
%D 2003
%J Journal of Multivariate Analysis
%K Graphical models
%N 2
%P 387 - 402
%R DOI: 10.1016/S0047-259X(02)00067-2
%T Bayesian networks for discrete multivariate data: an algebraic approach
to inference
%U http://www.sciencedirect.com/science/article/B6WK9-4806GJ3-9/2/009b39fdce41806005c14e6072047f04
%V 84
%X In this paper we demonstrate how Gr�bner bases and other algebraic
techniques can be used to explore the geometry of the probability
space of Bayesian networks with hidden variables. These techniques
employ a parametrisation of Bayesian network by moments rather than
conditional probabilities. We show that whilst Gr�bner bases help
to explain the local geometry of these spaces a complimentary analysis,
modelling the positivity of probabilities, enhances and completes
the geometrical picture. We report some recent geometrical results
in this area and discuss a possible general methodology for the analyses
of such problems.
@article{Smith2003b,
abstract = {In this paper we demonstrate how Gr�bner bases and other algebraic
techniques can be used to explore the geometry of the probability
space of Bayesian networks with hidden variables. These techniques
employ a parametrisation of Bayesian network by moments rather than
conditional probabilities. We show that whilst Gr�bner bases help
to explain the local geometry of these spaces a complimentary analysis,
modelling the positivity of probabilities, enhances and completes
the geometrical picture. We report some recent geometrical results
in this area and discuss a possible general methodology for the analyses
of such problems.},
added-at = {2009-09-12T19:19:34.000+0200},
author = {Smith, J. Q. and Croft, J.},
biburl = {https://www.bibsonomy.org/bibtex/26b67d0c38b4f161a3dafff5ae37507ea/mozaher},
doi = {DOI: 10.1016/S0047-259X(02)00067-2},
file = {:Smith2003b.pdf:PDF},
interhash = {1e3f16c303c09b0907671eae90211d45},
intrahash = {6b67d0c38b4f161a3dafff5ae37507ea},
issn = {0047-259X},
journal = {Journal of Multivariate Analysis},
keywords = {Graphical models},
number = 2,
owner = {Mozaherul Hoque},
pages = {387 - 402},
timestamp = {2009-09-12T19:19:42.000+0200},
title = {Bayesian networks for discrete multivariate data: an algebraic approach
to inference},
url = {http://www.sciencedirect.com/science/article/B6WK9-4806GJ3-9/2/009b39fdce41806005c14e6072047f04},
volume = 84,
year = 2003
}