Multi-scale analysis of graphs

• Applicants: Professor Dr. Gilles Blanchard; Professorin Dr. Ulrike von Luxburg
• Subject Area: Mathematics
• Term: from 2012 to 2020
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 197645397

The goal of this project is to develop statistical methodology for random graphs, taking hierarchical and multiscale structure into account.To this end we will on one hand extend the theory and methods developed in the first period for wavelets on graphs, to more general random graphs and statistical settings; and, on the other hand, develop new two-sample tests to test distribution equality of populations of random graphs, based on their local or global geometrical features.

• DFG Programme: Research Units
• Subproject of: FOR 1735: Structural Inference in Statistics: Adaptation and Efficiency