Path weights in undirected Markov random fields

Abstract

Probabilistic graphical models are an elegant framework which combines uncertainty and logicalstructure, through the use of graph theory, to compactly represent complex, real-worldphenomena. In graphical models a network is represented by a graph, that is a mathematicalobject where a set of points, called vertices, representing the variables of the system, canbe connected by lines or arrows, called edges, representing associations between variables.A fundamental concept in graph theory is that of path. One of the most common families of graphical models is represented by undirected graphical models (also called Markov random fields) represented by graphs where all the edges are undirected lines. The aim of this research project is to clarify the role played by the path weight in undirected graphical models providing a clear interpretation to the value they take. We believe that this will open the way to the use of these quantities in real-world applications.

Document Details

Document Type
DoD Grant Award
Publication Date
Feb 06, 2017
Source ID
FA95501710039

Entities

People

  • Alberto Roverato

Organizations

  • Air Force Office of Scientific Research
  • United States Air Force

Tags

Fields of Study

  • Computer science

Readers

  • Distributed Systems and Data Platform Development
  • Graph Algorithms and Convex Optimization.
  • Mathematical Modeling and Probability Theory.