Researchers regularly use graphical models, or graphs where each node is associated with a particular random variable. What types of algorithms offer the most efficient analyses of the relationship between different random variables represented in graphical models?

Alan Izenman offers an overview of the main algorithms used on graphical models today, as described by various prominent mathematicians. Specifically, Izenman focuses on the discussion of Discrete Markov Random Fields. There are many possible uses for Discrete Markov Random Fields, including analyses of Senate voting records, material science, melting arctic sea ice, computer vision and image analysis and restoring degraded photos.