Mean-field theory of Bayesian clustering

Dempster-Shafer clustering using Potts spin mean field theory

In this article we investigate a problem within Dempster±Shafer theory where 2 q À 1 pieces of evidence are clustered into q clusters by minimizing a metacon¯ict function, or equivalently, by minimizing the sum of weight of con¯ict over all clusters. Previously one of us developed a method based on a Hop®eld and Tank model. However, for very large problems we need a method with lower computatio...

Mean-field theory for clustering coefficients in Barabási-Albert networks.

We applied a mean-field approach to study clustering coefficients in Barabási-Albert (BA) networks. We found that the local clustering in BA networks depends on the node degree. Analytic results have been compared to extensive numerical simulations finding a very good agreement for nodes with low degrees. Clustering coefficient of a whole network calculated from our approach perfectly fits nume...

Embedded mean-field theory.

We introduce embedded mean-field theory (EMFT), an approach that flexibly allows for the embedding of one mean-field theory in another without the need to specify or fix the number of particles in each subsystem. EMFT is simple, is well-defined without recourse to parameters, and inherits the simple gradient theory of the parent mean-field theories. In this paper, we report extensive benchmarki...

Dynamical mean-field theory

Mean Field Implementation of Bayesian Ica

In this contribution we review the mean field approach to Bayesian independent component analysis (ICA) recently developed by the authors [1, 2]. For the chosen setting of additive Gaussian noise on the measured signal and Maximum Likelihood II estimation of the mixing matrix and the noise, the expected sufficient statistics are obtained from the two first posterior moments of the sources. Thes...