arXiv 1710.10903 · stat.ML cs.AI cs.LG
Graph Attention Networks
Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, Yoshua Bengio
Read on arXiv · arXiv non-exclusive license
- No reproduction recorded 8
Claims were extracted automatically by reducer-1 on 2026-11-12. The paper's license does not allow republishing it, so long claims are shortened to a quotation. Read the full text on arXiv.
We validate the proposed approach on four challenging benchmarks: Cora, Citeseer and Pubmed citation networks as well as an inductive protein-protein interactiondataset, achieving or matching state-of-the-art results that highlight the … (excerpt)
I NTRODUCTION · page 2 · Automated extraction, not reviewed · 0 reviewed reproductions
To evaluate the benefits of aggregating across the entire neighborhood, we further provide (as GraphSAGE∗) the best result we were able to achieve with GraphSAGE by just modifying its architecture … (excerpt)
R ESULTS · page 8 · Automated extraction, not reviewed · 0 reviewed reproductions
Our results successfully demonstrate state-of-the-art performance being achieved or matched across all four datasets—in concordance with our expectations, as per the discussion in Section 2.2.
R ESULTS · page 8 · Automated extraction, not reviewed · 0 reviewed reproductions
More specifically, we are able to improve upon GCNs by a margin of 1.5% and 1.6% on Cora and Citeseer, respectively, suggesting that assigning different weights to nodes of a … (excerpt)
R ESULTS · page 8 · Automated extraction, not reviewed · 0 reviewed reproductions
It is worth noting the improvements achieved on the PPI dataset: Our GAT model improves by 20.5% w.r.t. the best GraphSAGE result we were able to obtain, demonstrating that our … (excerpt)
R ESULTS · page 8 · Automated extraction, not reviewed · 0 reviewed reproductions
Furthermore, it improves by 3.9% w.r.t.
R ESULTS · page 8 · Automated extraction, not reviewed · 0 reviewed reproductions
We have presented graph attention networks (GATs), novel convolution-style neural networks that operate on graph-structured data, leveraging masked self-attentional layers.
C ONCLUSIONS · page 9 · Automated extraction, not reviewed · 0 reviewed reproductions
Our models leveraging attention have successfully achieved or matched state-of-the-art performance across four well-established node classification benchmarks, both transductive and inductive (especially, with completely unseen graphs used for testing).
C ONCLUSIONS · page 9 · Automated extraction, not reviewed · 0 reviewed reproductions