ProtoGNN: Prototype-Assisted Message Passing Framework for Non-Homophilous Graphs

Yanfei Dong, Mohammed Haroon Dupty, Lambert Deng, Yong Liang Goh, Wee Sun Lee

Keywords: Graph Neural Networks, Graph representation learning, Non-homophilous Graph, Heterophily, Non-homophily, Node Classification

TL;DR: Class prototype-assisted message passing framework for improving node representation learning on non-homophilous graphs

Abstract: Many well-known Graph Neural Network (GNN) models assume the underlying graphs are homophilous, where nodes share similar features and labels with their neighbours. They rely on message passing that iteratively aggregates neighbour's features and often suffer performance degradation on non-homophilous graphs where useful information is hardly available in the local neighbourhood. In addition, earlier studies show that in some cases GNNs are even outperformed by Multi-Layer Perceptron, indicating insufficient exploitation of node feature information. Motivated by the two limitations, we propose ProtoGNN, a novel message passing framework that augments existing GNNs by effectively combining node features with structural information. ProtoGNN learns multiple class prototypes for each class from raw node features with the slot-attention mechanism. These prototype representations are then transferred onto the structural node features with explicit message passing to all non-training nodes irrespective of distance. This form of message passing, from training nodes to class prototypes to non-training nodes, also serves as a shortcut that bypasses local graph neighbourhoods and captures global information. ProtoGNN is a generic framework which can be applied onto any of the existing GNN backbones to improve node representations when node features are strong and local graph information is scarce. We demonstrate through extensive experiments that ProtoGNN brings performance improvement to various GNN backbones and achieves state-of-the-art on several non-homophilous datasets.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Deep Learning and representational learning