Convergence Analysis of the Graph-Topology-Inference Kernel LMS Algorithm - ANR - Agence nationale de la recherche Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Convergence Analysis of the Graph-Topology-Inference Kernel LMS Algorithm

Ricardo Borsoi
Cédric Richard

Résumé

Identifying directed connectivity patterns from nodal measurements is an important problem in network analysis. Recent works proposed to leverage the performance and flexibility of strategies operating in reproducing kernel Hilbert spaces (RKHS) to model nonlinear interactions between network agents. Moreover, several applications require online and efficient solutions, which motivated the consideration of distributed adaptive learning strategies inspired by algorithms such as the kernel least mean square (KLMS). Despite showing good performance, a thorough theoretical understanding of the behavior of such algorithms is still missing. This makes applying them in practice challenging, especially because the setup of adaptive algorithms involves additional parameters like the step size and a dictionary of kernel functions. In this paper, we present a convergence analysis of the graph-topology-inference KLMS algorithm. Monte Carlo simulations demonstrate the accuracy of the theoretical models.
Fichier principal
Vignette du fichier
moscu2021convergence.pdf (1.7 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03350391 , version 1 (21-09-2021)

Identifiants

Citer

Mircea Moscu, Ricardo Borsoi, Cédric Richard. Convergence Analysis of the Graph-Topology-Inference Kernel LMS Algorithm. ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Jun 2021, Toronto, France. pp.5375-5379, ⟨10.1109/ICASSP39728.2021.9413693⟩. ⟨hal-03350391⟩
24 Consultations
68 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More