Graph Neural Networks with Heterophily
Preprint
Abstract
Graph Neural Networks (GNNs) have proven to be useful for many different practical applications. However, many existing GNN models have implicitly assumed homophily among the nodes connected in the graph, and therefore have largely overlooked the important setting of heterophily, where most connected nodes are from different classes. In this work, we propose a novel framework called CPGNN that generalizes GNNs for graphs with either homophily or heterophily. The proposed framework incorporates an interpretable compatibility matrix for modeling the heterophily or homophily level in the graph, which can be learned in an end-to-end fashion, enabling it to go beyond the assumption of strong homophily. Theoretically, we show that replacing the compatibility matrix in our framework with the identity (which represents pure homophily) reduces to GCN. Our extensive experiments demonstrate the effectiveness of our approach in more realistic and challenging experimental settings with significantly less training data compared to previous works: CPGNN variants achieve state-of-the-art results in heterophily settings with or without contextual node features, while maintaining comparable performance in homophily settings.Keywords
All Related Versions
This publication has 8 references indexed in Scilit:
- Two Sides of the Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks2022
- Multi-Scale attributed node embeddingJournal of Complex Networks, 2021
- Modeling polypharmacy side effects with graph convolutional networksBioinformatics, 2018
- Transforming Graph Data for Statistical Relational LearningJournal of Artificial Intelligence Research, 2012
- Cautious Inference in Collective Classification2007
- Birds of a Feather: Homophily in Social NetworksAnnual Review of Sociology, 2001
- Emergence of Scaling in Random NetworksScience, 1999
- Concerning nonnegative matrices and doubly stochastic matricesPacific Journal of Mathematics, 1967