ScholarGate
Assistent
Machine learning

Graph Attention Network

Graph Attention Network (GAT), introduceret af Veličković og kolleger i 2018, er en variant af grafneurale netværk, der lærer, hvor stor betydning der skal tildeles hver naboknude gennem en selvopmærksomhedsmekanisme (self-attention mechanism). På heterogene naboskaber og relationel klassifikation producerer den resultater, der er overlegne i forhold til grafkonvolutionelle netværk (GCN).

Åbn i MethodMindSnartVideoSnartDownload slides

Læs hele metoden

Kun for medlemmer

Log ind med en gratis konto for at læse dette afsnit.

Log ind

Method map

The neighbourhood of related methods — select a node to explore.

Graph Attention Network
Logistisk regressionRandom ForestRecurrent Neural NetworkXGBoostExponential Random Graph…Graph Convolutional Netw…Longformer / BigBirdMixture of ExpertsStokastisk blokmodelVisuel kontrastiv læring

Kilder

  1. Veličković, P. et al. (2018). Graph Attention Networks. ICLR. link ↗
  2. Brody, S. et al. (2022). How Attentive are Graph Attention Networks? ICLR. link ↗

Sådan citerer du denne side

ScholarGate. (2026, June 1). Graph Attention Network (GAT). ScholarGate. https://scholargate.app/da/deep-learning/graph-attention-network

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

Compare side by side →

Refereret af

Exponential Random Graph Model (ERGM / p*)Graph Convolutional Network (GCN)Longformer / BigBirdMixture of ExpertsStokastisk blokmodelVisuel kontrastiv læring

Har du fundet en fejl på denne side? Indberet den eller foreslå en rettelse →

ScholarGate — Graph Attention Network (Graph Attention Network (GAT)). Hentet 2026-06-15 fra https://scholargate.app/da/deep-learning/graph-attention-network · Datasæt: https://doi.org/10.5281/zenodo.20539026