Graph Learning Based on Total Variation Minimization

Peter Berger; Manfred Buchacher; Gabor Hannak; Gerald Matz

ICASSP 2018accepted0 citations

Graph Learning Based on Total Variation Minimization

Peter Berger, Manfred Buchacher, Gabor Hannak, Gerald Matz

Abstract

We consider the problem of learning the topology of a graph from a given set of smooth graph signals. We construct a weighted adjacency matrix that best explains the data in the sense of achieving the smallest graph total variation. For the case of noisy measurements of the graph signals we propose a scheme that simultaneously denoises the signals and learns the graph adjacency matrix. Our method allows for a direct control of the number of edges and of the weighted node degree. Numerical experiments demonstrate that our graph learning scheme is well suited for community detection.

BibTeX

@inproceedings{icassp2018_graphlearningbas,
  title = {Graph Learning Based on Total Variation Minimization},
  author = {Peter Berger and Manfred Buchacher and Gabor Hannak and Gerald Matz},
  booktitle = {ICASSP 2018},
  year = {2018}
}