Graph Attention Networks | Papers With Code

Browse State-of-the-Art
Datasets
Methods
More
Newsletter RC2022
About Trends Portals Libraries

Sign In

Subscribe to the PwC Newsletter

Stay informed on the latest trending ML papers with code, research developments, libraries, methods, and datasets.

Read previous issues

Join the community

You need to log in to edit.
You can create a new account if you don't have one.

Or, discuss a change on Slack.

TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Node Classification	Brazil Air-Traffic	GAT (Velickovic et al., 2018)	Accuracy	0.382	# 7
Node Classification	Chameleon (60%/20%/20% random splits)	GAT	1:1 Accuracy	63.9 ± 0.46	# 18
Node Classification on Non-Homophilic (Heterophilic) Graphs	Chameleon(60%/20%/20% random splits)	GAT	1:1 Accuracy	63.9 ± 0.46	# 18
Graph Classification	CIFAR10 100k	GAT	Accuracy (%)	65.48	# 10
Node Classification	Citeseer	GAT	Accuracy	72.5 ± 0.7%	# 43
Node Classification	Citeseer	GAT	Training Split	fixed 20 per node	# 3
Node Classification	Citeseer	GAT	Validation	YES	# 1
Node Classification	CiteSeer (0.5%)	GAT	Accuracy	38.2%	# 13
Node Classification	CiteSeer (1%)	GAT	Accuracy	46.5%	# 14
Node Classification	CiteSeer (60%/20%/20% random splits)	GAT	1:1 Accuracy	67.20 ± 0.46	# 31
Node Classification	CiteSeer with Public Split: fixed 20 nodes per class	GAT	Accuracy	72.5 ± 0.7%	# 26
Document Classification	Cora	GAT	Accuracy	83.0%	# 3
Node Classification	Cora	GAT	Accuracy	83.0% ± 0.7%	# 44
Node Classification	Cora	GAT	Training Split	fixed 20 per node	# 3
Node Classification	Cora	GAT	Validation	YES	# 1
Node Classification	Cora (0.5%)	GAT	Accuracy	41.4%	# 13
Node Classification	Cora (1%)	GAT	Accuracy	48.6%	# 14
Node Classification	Cora (3%)	GAT	Accuracy	56.8%	# 15
Node Classification	Cora (60%/20%/20% random splits)	GAT	1:1 Accuracy	76.70 ± 0.42	# 30
Node Classification	Cora with Public Split: fixed 20 nodes per class	GAT	Accuracy	83.0 ± 0.7%	# 21
Node Classification on Non-Homophilic (Heterophilic) Graphs	Cornell (60%/20%/20% random splits)	GAT	1:1 Accuracy	76.00 ± 1.01	# 26
Node Classification	Cornell (60%/20%/20% random splits)	GAT	1:1 Accuracy	76.00 ± 1.01	# 26
Node Classification on Non-Homophilic (Heterophilic) Graphs	Deezer-Europe	GAT	1:1 Accuracy	61.09±0.77	# 22
Node Classification	Europe Air-Traffic	GAT (Velickovic et al., 2018)	Accuracy	42.4	# 5
Node Classification	Film (60%/20%/20% random splits)	GAT	1:1 Accuracy	35.98 ± 0.23	# 26
Node Classification	Flickr	GAT (Velickovic et al., 2018)	Accuracy	0.359	# 8
Node Classification on Non-Homophilic (Heterophilic) Graphs	genius	GAT	1:1 Accuracy	55.80 ± 0.87	# 28
Node Classification	genius	GAT	Accuracy	55.80 ± 0.87	# 25
Skeleton Based Action Recognition	J-HMBD Early Action	GAT	10%	58.1	# 2
Graph Regression	Lipophilicity	GAT	RMSE	0.95	# 9
Graph Property Prediction	ogbg-code2	GAT	Test F1 score	0.1569 ± 0.0010	# 15
Graph Property Prediction	ogbg-code2	GAT	Validation F1 score	0.1442 ± 0.0017	# 14
Graph Property Prediction	ogbg-code2	GAT	Number of params	11030210	# 13
Graph Property Prediction	ogbg-code2	GAT	Ext. data	No	# 1
Node Property Prediction	ogbn-arxiv	GAT+label reuse+self KD	Test Accuracy	0.7416 ± 0.0008	# 23
Node Property Prediction	ogbn-arxiv	GAT+label reuse+self KD	Validation Accuracy	0.7514 ± 0.0004	# 28
Node Property Prediction	ogbn-arxiv	GAT+label reuse+self KD	Number of params	1441580	# 31
Node Property Prediction	ogbn-arxiv	GAT+label reuse+self KD	Ext. data	No	# 1
Node Property Prediction	ogbn-arxiv	GAT+label+reuse+topo loss	Test Accuracy	0.7399 ± 0.0012	# 29
Node Property Prediction	ogbn-arxiv	GAT+label+reuse+topo loss	Validation Accuracy	0.7513 ± 0.0009	# 29
Node Property Prediction	ogbn-arxiv	GAT+label+reuse+topo loss	Number of params	1441580	# 31
Node Property Prediction	ogbn-arxiv	GAT+label+reuse+topo loss	Ext. data	No	# 1
Node Property Prediction	ogbn-products	GAT with NeighborSampling	Test Accuracy	0.7945 ± 0.0059	# 45
Node Property Prediction	ogbn-products	GAT with NeighborSampling	Validation Accuracy	Please tell us	# 57
Node Property Prediction	ogbn-products	GAT with NeighborSampling	Number of params	751574	# 33
Node Property Prediction	ogbn-products	GAT with NeighborSampling	Ext. data	No	# 1
Node Property Prediction	ogbn-proteins	GAT + labels + node2vec	Test ROC-AUC	0.8711 ± 0.0007	# 8
Node Property Prediction	ogbn-proteins	GAT + labels + node2vec	Validation ROC-AUC	0.9217 ± 0.0011	# 9
Node Property Prediction	ogbn-proteins	GAT + labels + node2vec	Number of params	6360470	# 8
Node Property Prediction	ogbn-proteins	GAT + labels + node2vec	Ext. data	No	# 1
Node Classification	PATTERN 100k	GAT	Accuracy (%)	75.824	# 8
Node Classification	Penn94	GAT	Accuracy	81.53 ± 0.55	# 14
Node Classification on Non-Homophilic (Heterophilic) Graphs	Penn94	GAT	1:1 Accuracy	81.53 ± 0.55	# 14
Node Classification	PPI	GAT	F1	97.3	# 16
Node Classification	Pubmed	GAT	Accuracy	79.0	# 52
Node Classification	Pubmed	GAT	Training Split	fixed 20 per node	# 4
Node Classification	Pubmed	GAT	Validation	YES	# 1
Node Classification	PubMed (0.03%)	GAT	Accuracy	50.9%	# 12
Node Classification	PubMed (0.05%)	GAT	Accuracy	50.4%	# 13
Node Classification	PubMed (0.1%)	GAT	Accuracy	59.6%	# 13
Node Classification	PubMed (60%/20%/20% random splits)	GAT	1:1 Accuracy	83.28 ± 0.12	# 34
Node Classification	PubMed with Public Split: fixed 20 nodes per class	GAT	Accuracy	79.0%	# 22
Node Classification	Squirrel (60%/20%/20% random splits)	GAT	1:1 Accuracy	42.72 ± 0.33	# 22
Node Classification	Texas (60%/20%/20% random splits)	GAT	1:1 Accuracy	78.87 ± 0.86	# 31
Node Classification on Non-Homophilic (Heterophilic) Graphs	Texas(60%/20%/20% random splits)	GAT	1:1 Accuracy	78.87 ± 0.86	# 29

Edit Social Preview