New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction

Résumé

Graph autoencoders (GAE) and variational graph autoencoders (VGAE) emerged as powerful methods for link prediction (LP). Their performances are less impressive on community detection (CD), where they are often outperformed by simpler alternatives such as the Louvain method. It is still unclear to what extent one can improve CD with GAE and VGAE, especially in the absence of node features. It is moreover uncertain whether one could do so while simultaneously preserving good performances on LP in a multi-task setting. In this workshop paper, summarizing results from our journal publication (Salha-Galvan et al. 2022), we show that jointly addressing these two tasks with high accuracy is possible. For this purpose, we introduce a community-preserving message passing scheme, doping our GAE and VGAE encoders by considering both the initial graph and Louvain-based prior communities when computing embedding spaces. Inspired by modularity-based clustering, we further propose novel training and optimization strategies specifically designed for joint LP and CD. We demonstrate the empirical effectiveness of our approach, referred to as Modularity-Aware GAE and VGAE, on various real-world graphs.
Fichier principal
Vignette du fichier
NeurIPS_2022_Modularity.pdf (533.74 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04447637 , version 1 (08-02-2024)

Identifiants

Citer

Guillaume Salha-Galvan, Johannes Lutzeyer, George Dasoulas, Romain Hennequin, Michalis Vazirgiannis. New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction. NeurIPS New Frontiers in Graph Learning Workshop, Dec 2022, New Orleans, United States. ⟨hal-04447637⟩
4 Consultations
6 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More