Abstract:

Graphical models is a class of probabilistic models that can be understood pictorially using an interaction graph. At the core of these models is the concept of conditional independence; X is conditionally independent of Y given variable Z if by knowing Z, Y is irrelevant for X. Having access to such knowledge typically simplifies the model and hence the interaction graph.

For classification problems, the structure of the interaction graph may be of lesser importance, whereas the structure may be essential in problems where the goal is to understand the underlying phenomenon. The dissertation touches on outlier detection using graphical models, and how to leverage sparsity in the involved conditional probability tables in order to both decrease memory usage and speed up calculations during computations of posterior probabilities.

Graphical models can be used in wide range of applications, ranging from biology and genetics to engineering and social sciences.

Assessment Committee:       

Professor Steffen L. Lauritzen, University of Copenhagen

Professor Niels Richard Hansen, University of Copenhagen

Professor Thore Egeland, Norwegian University of Life Sciences

Supervisor:
Associate professor Torben Tvedebrink, Department of Mathematical Sciences, Aalborg University

Co-Supervisor:
Associate professor Poul Svante Eriksen, Department of Mathematical Sciences, Aalborg University

Co-Supervisor:
Professor Niels Morling, Department of Forensic Medicine, University of Copenhagen

The PhD defense will be hosted by Moderator professor Steffen L. Lauritzen. The lecture constitutes a 45 minutes presentation by Mads Lindskou followed by a short break and a discussion session with questions from the opponents and the auditorium.

After the defense the department host a small reception (cake, coffee and bubbles) at Skjernvej , room 5.123