On decomposable random graphs
نویسنده
چکیده
Abstract: Decomposable graphs are known for their tedious and complicated Markov update steps. Instead of modelling them directly, this work introduces a class of tree-dependent bipartite graphs that span the projective space of decomposable graphs. This is achieved through dimensionality expansion that causes the graph nodes to be conditionally independent given a latent tree. The Markov update steps are thus remarkably simplified. Structural modelling with tree-dependent bipartite graphs has additional benefits. For example, certain properties that are hardly attainable in the decomposable form are now easily accessible. Moreover, tree-dependent bipartite graphs can extract and model extra information related to sub-clustering dynamics, while currently known models for decomposable graphs do not. Properties of decomposable graphs are also transferable to the expanded dimension, such as the attractive likelihood factorization property. As a result of using the bipartite representation, tools developed for random graphs can be used. Hence, a framework for random tree-dependent bipartite graphs, thereupon for random decomposable graphs, is proposed.
منابع مشابه
Constructing vertex decomposable graphs
Recently, some techniques such as adding whiskers and attaching graphs to vertices of a given graph, have been proposed for constructing a new vertex decomposable graph. In this paper, we present a new method for constructing vertex decomposable graphs. Then we use this construction to generalize the result due to Cook and Nagel.
متن کاملA weaker structural Markov property for decomposable graph laws
We present a new kind of structural Markov property for probabilistic laws on decomposable graphs, prove the equivalence of an exponential family assumption, and discuss identifiability, modelling and computational implications. Some key words: conditional independence, graphical model, Markov random field, model determination, random graphs
متن کاملConditional Independence Models via Filtrations
We formulate a novel approach to infer conditional independence models or Markov structure of a multivariate distribution. Specifically, our objective is to place informative prior distributions over decomposable graphs and sample efficiently from the induced posterior distribution. The key idea we develop in this paper is a parametrization of decomposable hypergraphs using the geometry of poin...
متن کاملComplement of Special Chordal Graphs and Vertex Decomposability
In this paper, we introduce a subclass of chordal graphs which contains $d$-trees and show that their complement are vertex decomposable and so is shellable and sequentially Cohen-Macaulay.
متن کامل