Probabilistic Graphical Models

Catalog description

Probabilistic graphical models provide a flexible framework for analyzing large, complex, heterogeneous, and noisy data. They are the basis for state-of-the-art analysis methods in a wide variety of application domains, from autonomous robotics and computer vision to medical diagnosis and social networks. This course covers (a) representation, including Bayesian and Markov networks, (b) inference, both exact and approximate, and (c) estimation of both parameters and structure of graphical models.