Inferring Graph Signal Translations as Invariant Transformations for Classification Tasks - Equipe Algorithm Architecture Interactions Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Inferring Graph Signal Translations as Invariant Transformations for Classification Tasks

Résumé

The field of Graph Signal Processing (GSP) has proposed tools to generalize harmonic analysis to complex domains represented through graphs. Among these tools are translations, which are required to define many others. Most works propose to define translations using solely the graph structure (i.e. edges). Such a problem is ill-posed in general as a graph conveys information about neighborhood but not about directions. In this paper, we propose to infer translations as edge-constrained operations that make a supervised classification problem invariant using a deep learning framework. As such, our methodology uses both the graph structure and labeled signals to infer translations. We perform experiments with regular 2D images and abstract hyperlink networks to show the effectiveness of the proposed methodology in inferring meaningful translations for signals supported on graphs.
Fichier principal
Vignette du fichier
Graph_convolution_Eusipco.pdf (451.96 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03235669 , version 1 (25-05-2021)

Identifiants

Citer

Raphael Baena, Lucas Drumetz, Vincent Gripon. Inferring Graph Signal Translations as Invariant Transformations for Classification Tasks. EUSIPCO 2021: 29th European Signal Processing Conference, Aug 2021, Dublin, Ireland. ⟨10.23919/EUSIPCO54536.2021.9616010⟩. ⟨hal-03235669⟩
128 Consultations
108 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More