Simple Truncated SVD based Model for Node Classification on Heterophilic Graphs

24 Jun 2021  ·  Vijay Lingam, Rahul Ragesh, Arun Iyer, Sundararajan Sellamanickam ·

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.

PDF Abstract
No code implementations yet. Submit your code now

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Node Classification Actor HLP Concat Accuracy 34.59±1.32 # 12
Node Classification Chameleon HLP Concat Accuracy 77.48±0.80 # 3
Node Classification Cornell HLP Concat Accuracy 84.05±4.67 # 10
Node Classification Crocodile HLP Concat Accuracy 55.87±1.25 # 2
Node Classification Squirrel HLP Concat Accuracy 74.17±1.83 # 1
Node Classification Texas HLP Concat Accuracy 87.57 ± 5.44 # 7
Node Classification Wisconsin HLP Concat Accuracy 86.67±4.22 # 9


No methods listed for this paper. Add relevant methods here