When do GNNs work: Understanding and improving neighborhood aggregation

Yiqing Xie, Sha Li, Carl Yang, Raymond Chi Wing Wong, Jiawei Han

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Graph Neural Networks (GNNs) have been shown to be powerful in a wide range of graph-related tasks. While there exist various GNN models, a critical common ingredient is neighborhood aggregation, where the embedding of each node is updated by referring to the embedding of its neighbors. This paper aims to provide a better understanding of this mechanism by asking the following question: Is neighborhood aggregation always necessary and beneficial? In short, the answer is no. We carve out two conditions under which neighborhood aggregation is not helpful: (1) when a node's neighbors are highly dissimilar and (2) when a node's embedding is already similar to that of its neighbors. We propose novel metrics that quantitatively measure these two circumstances and integrate them into an Adaptive-layer module. Our experiments show that allowing for node-specific aggregation degrees have significant advantage over current GNNs.

Original languageEnglish (US)
Title of host publicationProceedings of the 29th International Joint Conference on Artificial Intelligence, IJCAI 2020
EditorsChristian Bessiere
PublisherInternational Joint Conferences on Artificial Intelligence
Pages1303-1309
Number of pages7
ISBN (Electronic)9780999241165
StatePublished - 2020
Event29th International Joint Conference on Artificial Intelligence, IJCAI 2020 - Yokohama, Japan
Duration: Jan 1 2021 → …

Publication series

NameIJCAI International Joint Conference on Artificial Intelligence
Volume2021-January
ISSN (Print)1045-0823

Conference

Conference29th International Joint Conference on Artificial Intelligence, IJCAI 2020
CountryJapan
CityYokohama
Period1/1/21 → …

ASJC Scopus subject areas

  • Artificial Intelligence

Fingerprint Dive into the research topics of 'When do GNNs work: Understanding and improving neighborhood aggregation'. Together they form a unique fingerprint.

Cite this