Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns

Benchmark Model Rank Results
node-classification-on-actorWRGAT#26Accuracy: 36.53 ± 0.77
node-classification-on-chameleonWRGAT#32Accuracy: 65.24 ± 0.87
node-classification-on-citeseer-48-32-20WRGAT#151:1 Accuracy: 76.81 ± 1.89
node-classification-on-cora-48-32-20-fixedWRGAT#61:1 Accuracy: 88.20 ± 2.26
node-classification-on-cornellWRGAT#28Accuracy: 81.62 ± 3.90
node-classification-on-non-homophilic-10WRGAT#131:1 Accuracy: 36.53 ± 0.77
node-classification-on-non-homophilic-11WRGAT#201:1 Accuracy: 65.24 ± 0.87
node-classification-on-non-homophilic-12WRGAT#201:1 Accuracy: 48.85 ± 0.78
node-classification-on-non-homophilic-13WRGAT#231:1 Accuracy: 74.32 ± 0.53
node-classification-on-non-homophilic-7WRGAT#181:1 Accuracy: 81.62 ±3.90
node-classification-on-non-homophilic-8WRGAT#141:1 Accuracy: 86.98 ± 3.78
node-classification-on-non-homophilic-9WRGAT#151:1 Accuracy: 83.62 ± 5.50
node-classification-on-penn94WRGAT#27Accuracy: 74.32 ± 0.53
node-classification-on-pubmed-48-32-20-fixedWRGAT#181:1 Accuracy: 88.52 ± 0.92
node-classification-on-squirrelWRGAT#30Accuracy: 48.85 ± 0.78
node-classification-on-texasWRGAT#28Accuracy: 83.62 ± 5.50