GAT (Velickovic et al., 2018)

Reported on 5 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Graphs5 results

Node ClassificationonWiki-Vote
Accuracy· 2017-10-30
59.4
best: 99.8 (DEMO-Net(weight))
SOTA
Graph Attention Networks arXiv:1710.10903
Node ClassificationonBrazil Air-Traffic
Accuracy· 2017-10-30
0.382
best: 0.516 (GCN_cheby (Kipf and Welling, 2017))
Graph Attention Networks arXiv:1710.10903
Node ClassificationonEurope Air-Traffic
Accuracy· 2017-10-30
42.4
best: 46 (GCN_cheby (Kipf and Welling, 2017))
Graph Attention Networks arXiv:1710.10903
Node ClassificationonFlickr
Accuracy· 2017-10-30
0.359
best: 0.682 (GCN+GAugM (Zhao et al., 2021))
Graph Attention Networks arXiv:1710.10903
Node ClassificationonUSA Air-Traffic
Accuracy· 2017-10-30
58.5
best: 64.7 (DEMO-Net(weight))
Graph Attention Networks arXiv:1710.10903