Adaptive Universal Generalized PageRank Graph Neural Network
This addresses the challenge of designing universal GNNs that work well across different graph types, offering a solution for applications in graph data processing where label patterns vary, though it is incremental as it builds on existing GPR and GNN methods.
The paper tackles the problem of optimally integrating node features and graph topology in graph neural networks (GNNs) for node classification, regardless of homophily or heterophily assumptions, by introducing an adaptive Generalized PageRank (GPR) GNN that learns weights to adjust to label patterns and avoid feature over-smoothing, resulting in significant performance improvements on synthetic and benchmark datasets.
In many important graph data processing applications the acquired information includes both node features and observations of the graph topology. Graph neural networks (GNNs) are designed to exploit both sources of evidence but they do not optimally trade-off their utility and integrate them in a manner that is also universal. Here, universality refers to independence on homophily or heterophily graph assumptions. We address these issues by introducing a new Generalized PageRank (GPR) GNN architecture that adaptively learns the GPR weights so as to jointly optimize node feature and topological information extraction, regardless of the extent to which the node labels are homophilic or heterophilic. Learned GPR weights automatically adjust to the node label pattern, irrelevant on the type of initialization, and thereby guarantee excellent learning performance for label patterns that are usually hard to handle. Furthermore, they allow one to avoid feature over-smoothing, a process which renders feature information nondiscriminative, without requiring the network to be shallow. Our accompanying theoretical analysis of the GPR-GNN method is facilitated by novel synthetic benchmark datasets generated by the so-called contextual stochastic block model. We also compare the performance of our GNN architecture with that of several state-of-the-art GNNs on the problem of node-classification, using well-known benchmark homophilic and heterophilic datasets. The results demonstrate that GPR-GNN offers significant performance improvement compared to existing techniques on both synthetic and benchmark data.