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.
In this way, we address several key challenges of spectral-based graph neural networks
simultaneously, and make our model readily applicable to inductive as well
as transductive problems. Our GAT models have achieved or matched state-of-theart
results across four established transductive and inductive graph benchmarks:
the Cora, Citeseer and Pubmed citation network datasets, as well as a proteinprotein
interaction dataset (wherein test graphs remain unseen during training).
%0 Journal Article
%1 journals/corr/abs-1710-10903
%A Velickovic, Petar
%A Cucurull, Guillem
%A Casanova, Arantxa
%A Romero, Adriana
%A Liò, Pietro
%A Bengio, Yoshua
%D 2017
%J CoRR
%K graph
%T Graph Attention Networks.
%U http://dblp.uni-trier.de/db/journals/corr/corr1710.html#abs-1710-10903
%V abs/1710.10903
%X 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.
In this way, we address several key challenges of spectral-based graph neural networks
simultaneously, and make our model readily applicable to inductive as well
as transductive problems. Our GAT models have achieved or matched state-of-theart
results across four established transductive and inductive graph benchmarks:
the Cora, Citeseer and Pubmed citation network datasets, as well as a proteinprotein
interaction dataset (wherein test graphs remain unseen during training).
@article{journals/corr/abs-1710-10903,
abstract = {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.
In this way, we address several key challenges of spectral-based graph neural networks
simultaneously, and make our model readily applicable to inductive as well
as transductive problems. Our GAT models have achieved or matched state-of-theart
results across four established transductive and inductive graph benchmarks:
the Cora, Citeseer and Pubmed citation network datasets, as well as a proteinprotein
interaction dataset (wherein test graphs remain unseen during training).},
added-at = {2019-08-12T10:13:26.000+0200},
author = {Velickovic, Petar and Cucurull, Guillem and Casanova, Arantxa and Romero, Adriana and Liò, Pietro and Bengio, Yoshua},
biburl = {https://www.bibsonomy.org/bibtex/2bfbede338b93a29d03171a36e5fdada2/tobias.koopmann},
ee = {http://arxiv.org/abs/1710.10903},
interhash = {198ba105f462abe0d5d06890adc70a90},
intrahash = {bfbede338b93a29d03171a36e5fdada2},
journal = {CoRR},
keywords = {graph},
timestamp = {2019-09-04T10:54:06.000+0200},
title = {Graph Attention Networks.},
url = {http://dblp.uni-trier.de/db/journals/corr/corr1710.html#abs-1710-10903},
volume = {abs/1710.10903},
year = 2017
}