GNN Transformation Framework for Improving Efficiency and Scalability
This work addresses scalability issues for researchers and practitioners using GNNs on large-scale graphs, though it is incremental as it builds on existing precomputation methods.
The paper tackles the problem of scaling Graph Neural Networks (GNNs) to large graphs by proposing a framework that transforms non-scalable GNNs into efficient precomputation-based versions, achieving faster training times while maintaining competitive accuracy with state-of-the-art GNNs.
We propose a framework that automatically transforms non-scalable GNNs into precomputation-based GNNs which are efficient and scalable for large-scale graphs. The advantages of our framework are two-fold; 1) it transforms various non-scalable GNNs to scale well to large-scale graphs by separating local feature aggregation from weight learning in their graph convolution, 2) it efficiently executes precomputation on GPU for large-scale graphs by decomposing their edges into small disjoint and balanced sets. Through extensive experiments with large-scale graphs, we demonstrate that the transformed GNNs run faster in training time than existing GNNs while achieving competitive accuracy to the state-of-the-art GNNs. Consequently, our transformation framework provides simple and efficient baselines for future research on scalable GNNs.