Renormalized Graph Representations for Node Classification
This addresses node classification in graph neural networks, but it is incremental as it builds on existing renormalization theory to enhance performance.
The paper tackled the problem of node classification by analyzing the effect of using coarse-grained graph resolutions from Laplacian renormalization group theory, specifically at the characteristic scale of mesoscopic structures, and found that models with access to both original and characteristic scale graphs achieved statistically significant improvements in test accuracy.
Graph neural networks process information on graphs represented at a given resolution scale. We analyze the effect of using different coarse-grained graph resolutions, obtained through the Laplacian renormalization group theory, on node classification tasks. At the theory's core is grouping nodes connected by significant information flow at a given time scale. Representations of the graph at different scales encode interaction information at different ranges. We specifically experiment using representations at the characteristic scale of the graph's mesoscopic structures. We provide the models with the original graph and the graph represented at the characteristic resolution scale and compare them to models that can only access the original graph. Our results showed that models with access to both the original graph and the characteristic scale graph can achieve statistically significant improvements in test accuracy.