Relation of Loopy Sum-Product Algorithm and Rank-1 Tensor Approximation

Project Type: Master/Diploma Thesis
Student: Carlos Eduardo Cancino Chacon

Short Description

Loopy belief propagation is well established for performing approximate inference in loopy graphs. Recently, we showed that the belief propagation equations can be merged into a single fixed point equation using methods borrowed from tensor analysis. The aim of this thesis is to investigate the convergence behaviour of this fixed point equation.



Your Profile/Requirements

The candidate should be interested in machine learning, applied mathematics/statistics, Matlab programming, and algorithms. Interested candidates are encouraged to ask for further information. Additionally, the supervision of own projects in one of the above mention fields is possible.


Franz Pernkopf ( or 0316/873 4436)