Graph Attention Networks ​

← Back to topic

Authors: Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, Yoshua Bengio
Year: 2018
Journal: ICLR
DOI: 10.48550/arXiv.1710.10903
Publisher: https://arxiv.org/abs/1710.10903

Keywords: gat, gnn

Abstract ​

We present graph attention networks GATs novel neural network architectures that operate on graph-structured data leveraging masked self-attention layers.

Cite this paper ​

@misc{gat2018,
  title  = {Graph Attention Networks},
  author = {Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, Yoshua Bengio},
  year   = {2018},
  journal = {ICLR},
  doi    = {10.48550/arXiv.1710.10903},
  url    = {https://doi.org/10.48550/arXiv.1710.10903},
}

Source files ​

• metadata.json
• paper.bib