A supervised edge-sampling sparsifier that keeps only 20% of edges can match or beat the full graph for GNN node classification, with especially large gains on heterophilic graphs.
Demystifying graph sparsification algorithms in graph properties preservation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SGS-GNN: A Supervised Graph Sparsification method for Graph Neural Networks
A supervised edge-sampling sparsifier that keeps only 20% of edges can match or beat the full graph for GNN node classification, with especially large gains on heterophilic graphs.