Improving Graph Neural Networks at Scale: Combining Approximate PageRank and CoreRank - Archive ouverte HAL Access content directly
Conference Papers Year : 2022

Improving Graph Neural Networks at Scale: Combining Approximate PageRank and CoreRank

Abstract

Graph Neural Networks (GNNs) have achieved great successes in many learning tasks performed on graph structures. Nonetheless, to propagate information GNNs rely on a message passing scheme which can become prohibitively expensive when working with industrial-scale graphs. Inspired by the PPRGo model, we propose the CorePPR model, a scalable solution that utilises a learnable convex combination of the approximate personalised PageRank and the CoreRank to diffuse multi-hop neighbourhood information in GNNs. Additionally, we incorporate a dynamic mechanism to select the most influential neighbours for a particular node which reduces training time while preserving the performance of the model. Overall, we demonstrate that CorePPR outperforms PPRGo, particularly on large graphs where selecting the most influential nodes is particularly relevant for scalability.
Fichier principal
Vignette du fichier
ImprovingGraphNeuralNetworksAtScaleCombiningApproximatePageRankAndCoreRank.pdf (302.26 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03838863 , version 1 (03-11-2022)

Licence

Identifiers

  • HAL Id : hal-03838863 , version 1

Cite

Ariel R. Ramos Vela, Johannes F. Lutzeyer, Anastasios Giovanidis, Michalis Vazirgiannis. Improving Graph Neural Networks at Scale: Combining Approximate PageRank and CoreRank. New Frontiers in Graph Learning (GLFrontiers) NeurIPS Workshop 2022, Dec 2022, New Orleans, United States. ⟨hal-03838863⟩
78 View
63 Download

Share

Gmail Mastodon Facebook X LinkedIn More