NPA: Improving Large-scale Graph Neural Networks with Non-parametric Attention
Open Access
- 9 June 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 414-427
Abstract
Recent works show great interest in designing Graph Neural Networks (GNNs) that scale to large graphs. While previous work focuses on designing advanced sampling techniques for existing GNNs, the design of non-parametric GNNs, an orthogonal direction for scalable performance, has aroused lots of concerns recently. For example, nearly all top solutions in the Open Graph Benchmark leaderboard are non-parametric GNNs. Despite their high predictive performance and scalability, non-parametric GNNs still face two limitations. First, due to the propagation of over-smoothed features, they suffer from severe performance degradation along with the propagation depth. More importantly, they only consider the graph structure and ignore the feature influence during the non-parametric propagation, leading to sub-optimal propagated features. To address these limitations, we present non-parametric attention (NPA), a plug-and-play module that is compatible with non-parametric GNNs, to get scalable and deep GNNs simultaneously. We have deployed NPA in Tencent with the Angel platform, and we further evaluate NPA on both real-world datasets and large-scale industrial datasets. Experimental results on seven homophilic graphs (including the industrial Tencent Video graph) and five heterophilic graphs demonstrate NPA enjoys high performance -- achieves large performance gain over existing non-parametric GNNs, deeper architecture -- improves non-parametric GNNs with large model depth, and high scalability -- can support large-scale graphs with low time costs. Notably, it achieves state-of-the-art performance on the large ogbn-papers100M dataset.Keywords
Funding Information
- NSFC (22B2037, U23B2048)
This publication has 20 references indexed in Scilit:
- Scapin: Scalable Graph Structure Perturbation by Augmented Influence MaximizationProceedings of the ACM on Management of Data, 2023
- Deep graph learning for spatially-varying indoor lighting predictionScience China Information Sciences, 2023
- Graph Attention Multi-Layer Perceptron2022
- PaSca: A Graph Neural Architecture Search System under the Scalable Paradigm2022
- GBK-GNN: Gated Bi-Kernel Graph Neural Networks for Modeling Both Homophily and Heterophily2022
- Graph Neural Networks for Recommender System2022
- A Survey of Machine Learning for Computer Architecture and SystemsACM Computing Surveys, 2022
- Point-GNN: Graph Neural Network for 3D Object Detection in a Point Cloud2020
- Microsoft Academic Graph: When experts are not enoughQuantitative Science Studies, 2020
- Cluster-GCN2019