Robust Method for Network Topology Identification Under Structural Equation Model

We present a robust method to infer network topology in the presence of outliers from given observations at nodes under the structural equation model. We introduce auxiliary matrices modeling Gaussian noise and sparse outliers. The topology identification task is cast as a minimization problem of th...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2024 IEEE 34th International Workshop on Machine Learning for Signal Processing (MLSP) S. 1 - 6
Hauptverfasser: Yoshida, Kohei, Yukawa, Masahiro
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 22.09.2024
Schlagworte:
graph learning
Mathematical models
Minimization methods
Network topology
operator splitting algorithm
outlier robustness
Pollution measurement
Robustness
Signal processing algorithms
Simulation
Sparse matrices
structural equation model
Topology
Vectors
ISSN:2161-0371
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We present a robust method to infer network topology in the presence of outliers from given observations at nodes under the structural equation model. We introduce auxiliary matrices modeling Gaussian noise and sparse outliers. The topology identification task is cast as a minimization problem of the sum of three terms under constraints involving a bilinear form: (i) the squared Frobenius norm of the noise matrix, (ii) the \ell_{1} norm of the adjacency matrix, and (iii) a weakly-convex sparsity-promoting function (the minimax concave penalty) of the outlier matrix. The problem is reformulated into an unconstrained optimization problem by introducing a linear operator, and an efficient alternating minimization method is presented. Simulation results show remarkable robustness of the proposed method.
ISSN:2161-0371
DOI:10.1109/MLSP58920.2024.10734807