Matz, G., Verardo, C., & Dittrich, T. (2023). Efficient Learning of Balanced Signature Graphs. In IEEE Proc. International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE ICASSP 2023, Rhodos, Greece. https://doi.org/10.1109/ICASSP49357.2023.10095989
E389-03 - Forschungsbereich Signal Processing E389 - Institute of Telecommunications
-
Published in:
IEEE Proc. International Conference on Acoustics, Speech and Signal Processing (ICASSP)
-
ISBN:
978-1-7281-6327-7
-
Date (published):
1-Jan-2023
-
Event name:
IEEE ICASSP 2023
-
Event date:
4-Jun-2023 - 10-Jun-2023
-
Event place:
Rhodos, Greece
-
Number of Pages:
5
-
Keywords:
Graph Signal Processing, Clustering Algorithms, Classification Algorithms
-
Abstract:
The novel concept of signature graphs extends signed graphs by admitting multiple types of partial similarity/agreement or dissimilarity/disagreement. Extending the concept of balancedness to signature graphs yields an explicit and efficient basis for multi-class clustering and classification. Contrary to existing two-stage approaches that consist of graph learning followed by graph clustering, we propose a one-step procedure that directly learns a perfectly clustered graph. We describe the algorithmic constituents for our approach and illustrate its superiority via numerical simulations.
