TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Node Classification	CiteSeer (0.5%)	GAT	Accuracy	38.2%	# 13
Node Classification	CiteSeer (1%)	GAT	Accuracy	46.5%	# 13
Node Classification	CiteSeer with Public Split: fixed 20 nodes per class	GAT	Accuracy	72.5%	# 8
Document Classification	Cora	GAT	Accuracy	83.0%	# 3
Node Classification	Cora (0.5%)	GAT	Accuracy	41.4%	# 12
Node Classification	Cora (1%)	GAT	Accuracy	48.6%	# 13
Node Classification	Cora (3%)	GAT	Accuracy	56.8%	# 14
Node Classification	Cora with Public Split: fixed 20 nodes per class	GAT	Accuracy	83.0%	# 9
Skeleton Based Action Recognition	J-HMBD Early Action	GAT	10%	58.1	# 2
Graph Regression	Lipophilicity	GAT	RMSE	0.950	# 8
Node Classification	PPI	GAT	F1	97.3	# 11
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 with Public Split: fixed 20 nodes per class	GAT	Accuracy	79.0%	# 8

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/skeleton-based-action-recognition-on-j-hmbd)](https://paperswithcode.com/sota/skeleton-based-action-recognition-on-j-hmbd?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/document-classification-on-cora)](https://paperswithcode.com/sota/document-classification-on-cora?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-citeseer-with-public)](https://paperswithcode.com/sota/node-classification-on-citeseer-with-public?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/graph-regression-on-lipophilicity)](https://paperswithcode.com/sota/graph-regression-on-lipophilicity?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-pubmed-with-public)](https://paperswithcode.com/sota/node-classification-on-pubmed-with-public?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-cora-with-public-split)](https://paperswithcode.com/sota/node-classification-on-cora-with-public-split?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-ppi)](https://paperswithcode.com/sota/node-classification-on-ppi?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-cora-05)](https://paperswithcode.com/sota/node-classification-on-cora-05?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-pubmed-003)](https://paperswithcode.com/sota/node-classification-on-pubmed-003?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-citeseer-05)](https://paperswithcode.com/sota/node-classification-on-citeseer-05?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-citeseer-1)](https://paperswithcode.com/sota/node-classification-on-citeseer-1?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-cora-1)](https://paperswithcode.com/sota/node-classification-on-cora-1?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-pubmed-005)](https://paperswithcode.com/sota/node-classification-on-pubmed-005?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-pubmed-01)](https://paperswithcode.com/sota/node-classification-on-pubmed-01?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-citeseer)](https://paperswithcode.com/sota/node-classification-on-citeseer?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-cora-3)](https://paperswithcode.com/sota/node-classification-on-cora-3?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-cora)](https://paperswithcode.com/sota/node-classification-on-cora?p=graph-attention-networks)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/graph-attention-networks/node-classification-on-pubmed)](https://paperswithcode.com/sota/node-classification-on-pubmed?p=graph-attention-networks)`

Graph Attention Networks

ICLR 2018 • Petar Veličković • Guillem Cucurull • Arantxa Casanova • Adriana Romero • Pietro Liò • Yoshua Bengio

We present graph attention networks (GATs), novel neural network architectures that operate on graph-structured data, leveraging masked self-attentional layers to address the shortcomings of prior methods based on graph convolutions or their approximations. By stacking layers in which nodes are able to attend over their neighborhoods' features, we enable (implicitly) specifying different weights to different nodes in a neighborhood, without requiring any kind of costly matrix operation (such as inversion) or depending on knowing the graph structure upfront... (read more)

PDF Abstract

Code

Add Remove

PetarV-/GAT

1,171

Diego999/pyGAT

699

danielegrattarola/keras-gat

210

shenweichen/GraphNeuralNetwork

125

HazyResearch/hgcn

See all 26 implementations

Tasks

Add Remove

DOCUMENT CLASSIFICATION

GRAPH EMBEDDING

GRAPH REGRESSION

LINK PREDICTION

NODE CLASSIFICATION

SKELETON BASED ACTION RECOGNITION

Evaluation Results from the Paper

Add Remove

#2 best model for Skeleton Based Action Recognition on J-HMBD Early Action

Get a GitHub badge

TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK	COMPARE
Node Classification	CiteSeer (0.5%)	GAT	Accuracy	38.2%	# 13	See all
Node Classification	CiteSeer (1%)	GAT	Accuracy	46.5%	# 13	See all
Node Classification	CiteSeer with Public Split: fixed 20 nodes per class	GAT	Accuracy	72.5%	# 8	See all
Document Classification	Cora	GAT	Accuracy	83.0%	# 3	See all
Node Classification	Cora (0.5%)	GAT	Accuracy	41.4%	# 12	See all
Node Classification	Cora (1%)	GAT	Accuracy	48.6%	# 13	See all
Node Classification	Cora (3%)	GAT	Accuracy	56.8%	# 14	See all
Node Classification	Cora with Public Split: fixed 20 nodes per class	GAT	Accuracy	83.0%	# 9	See all
Skeleton Based Action Recognition	J-HMBD Early Action	GAT	10%	58.1	# 2	See all
Graph Regression	Lipophilicity	GAT	RMSE	0.950	# 8	See all
Node Classification	PPI	GAT	F1	97.3	# 11	See all
Node Classification	PubMed (0.03%)	GAT	Accuracy	50.9%	# 12	See all
Node Classification	PubMed (0.05%)	GAT	Accuracy	50.4%	# 13	See all
Node Classification	PubMed (0.1%)	GAT	Accuracy	59.6%	# 13	See all
Node Classification	PubMed with Public Split: fixed 20 nodes per class	GAT	Accuracy	79.0%	# 8	See all

Graph Attention Networks

Code Edit Add Remove

Tasks Edit Add Remove

Evaluation Results from the Paper Edit Add Remove

Code

Add Remove

Tasks

Add Remove

Evaluation Results from the Paper

Add Remove