BAYESIAN NODE CLASSIFICATION FOR NOISY GRAPHS - Equipe Communications numériques Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

BAYESIAN NODE CLASSIFICATION FOR NOISY GRAPHS

Résumé

Graph neural networks (GNN) have been recognized as powerful tools for learning representations in graph structured data. The key idea is to propagate and aggregate information along edges of the given graph. However, little work has been done to analyze the effect of noise on their performance. By conducting a number of simulations, we show that GNN are very sensitive to the graph noise. We propose a graphassisted Bayesian node classifier which takes into account the degree of impurity of the graph, and show that it consistently outperforms GNN based classifiers on benchmark datasets, particularly when the degree of impurity is moderate to high.
Fichier principal
Vignette du fichier
ssp2021.pdf (281.31 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03291075 , version 1 (19-07-2021)

Identifiants

  • HAL Id : hal-03291075 , version 1

Citer

Hakim Hafidi, Mounir Ghogho, Philippe Ciblat, Ananthram Swami. BAYESIAN NODE CLASSIFICATION FOR NOISY GRAPHS. IEEE Statistical Signal Processing Workshop (SSP), 2021, Rio de Janeiro (virtual), Brazil. ⟨hal-03291075⟩
52 Consultations
191 Téléchargements

Partager

Gmail Facebook X LinkedIn More