Simple Truncated SVD based Model for Node Classification on Heterophilic Graphs

  • Vijay Lingam ,
  • Rahul Ragesh ,
  • ,
  • Sundararajan Sellamanickam

Knowledge Discovery and Data Mining Workshop on Deep Learning on Graphs: Method and Applications (DLG-KDD) |

Graph Neural Networks (GNNs) have shown excellent performance on graphs that exhibit strong homophily with respect to the node labels i.e. connected nodes have same labels. However, they perform poorly on heterophilic graphs. Recent approaches have typically modified aggregation schemes, designed adaptive graph filters, etc. to address this limitation. In spite of this, the performance on heterophilic graphs can still be poor. We propose a simple alternative method that exploits Truncated Singular Value Decomposition (TSVD) of topological structure and node features. Our approach achieves up to ~30% improvement in performance over state-of-the-art methods on heterophilic graphs. This work is an early investigation into methods that differ from aggregation based approaches. Our experimental results suggest that it might be important to explore other alternatives to aggregation methods for heterophilic setting.