Properties of Kikuchi approximations constructed from clique based decompositions

Download: PDF.

“Properties of Kikuchi approximations constructed from clique based decompositions” by Roberto Santana, P. Larrañaga, and J. A. Lozano, Department of Computer Science and Artificial Intelligence. University of the Basque Country technical report EHU-KZAA-IK-2/05, Apr. 2005.

Abstract

Kikuchi approximations constructed from clique-based decompositions can be used to calculate suitable approximations of probability distributions. They can be applied in domains such as probabilistic modeling, supervised and unsupervised classi cation, and evolutionary algorithms. This paper introduces a number of properties of these approximations. Pairwise and local Markov properties of the Kikuchi approximations are proved. We prove that, even if the global Markov property is not satisfied in the general case, it is possible to decompose the Kikuchi approximation in the product of local Kikuchi approximations defined on a decomposition of the graph. Partial Kikuchi approximations are introduced. Additionally, the paper clarifies the place of clique-based decompositions in relation to other techniques inspired by methods from statistical physics, and discusses the application of the results introduced in the paper for the conception of Kikuchi approximation learning algorithms.

Download: PDF.

BibTeX entry:

@techreport{Santana_et_al:2005a,
   author = {Roberto Santana and P. Larra{\~n}aga and J. A. Lozano},
   title = {Properties of {K}ikuchi approximations constructed from clique
	based decompositions},
   institution = {Department of Computer Science and Artificial
	Intelligence, University of the Basque Country},
   number = {EHU-KZAA-IK-2/05},
   month = apr,
   year = {2005}
}

(This webpage was created with bibtex2web.)

Back to Roberto Santana publications.