Graphical Models for Discrete and Continuous Data (306549)Johannes Lederer, Ruhr-University Bochum
Noah Simon, University of Washington
*Rui Zhuang, University of Washington
Keywords: Graphical Models, Maximum Likelihood, Inference
We introduce a general framework for undirected graph- ical models. It generalizes Gaussian graphical models to a wide range of continuous, discrete, and combinations of different types of data. We also show that the models in the framework, called exponential trace models, are amenable to efficient estimation and inference based on maximum likelihood. As a consequence, we expect applications to a large variety of multivariate data that have correlated coordinates.