Browsed by
Tag: Yedidia

Gibbs Free Energy, Belief Propagation, and Markov Random Fields

Gibbs Free Energy, Belief Propagation, and Markov Random Fields

Correspondence Between Free Energy, Belief Propagation, and Markov Random Field Models As a slight digression from previous posts – re-reading the paper by Yedidia et al. on this morning on Understanding Belief Propagation and its Generalizations – which explains the close connection between Belief Propagation (BP) methods and the Bethe approximation (a more generalized version of the simple bistate Ising model that I’ve been using) in statistical thermodynamics. The important point that Yedidia et al. make is that their work…

Read More Read More

Community Detection in Graphs

Community Detection in Graphs

Complexity and Graph Theory: A Brief Note Santo Fortunato has published an interesting and densly rich article, Community Detection in Graphs, in  Complexity (Inter-Wiley). This article is over 100 pages long, it is relatively complete, with numerous references and excellent figures. It is a bit surprising, however, that this extensive discussion misses one of the things that would seem to be most important in discussing graphs, and particularly, clusters within graphs: the stability of these clusters. That is; the theoretical basis for cluster…

Read More Read More

Graph Theory — Becoming "Organizing Framework"

Graph Theory — Becoming "Organizing Framework"

Something I’ve been noting — both on my own, and in conversations with Jenn Sleeman , who’s in touch with the academic world at UMBC — Graph theory is growing in centrality as a fundamental organizing framework for many current and emerging computational processes. Specifically, anything more complex than a simple “tuple” (RDF or OWL, etc.), needs to be matched against a graph or partial graph. One good “integrative” paper is Understanding Belief Propagation and its Generalizations by J.S. Yedidia,…

Read More Read More