.png){kind=icon}
返回
Multi-head multi-order graph attention networks
.png){kind=icon}
PRE
AI
DOI：10.1007/s10489-024-05601-z.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
The Graph Attention Network (GAT) is a type of graph neural network (GNN) that uses attention mechanisms to weigh the importance of nodes' neighbors, demonstrating flexibility and power in representation learning. However, GAT and its variants still face common challenges in GNNs, such as over-smoothing and over-squashing. To address this, we propose Multi-Head Multi-Order Graph Attention Networks (MHMOGAT) as an enhanced GAT layer. MHMOGAT is built based on multi-head attention and adjacency matrices of different orders, aiming to expand the receptive field of GAT to effectively capture long-distance dependencies. Moreover, Bayesian optimization is employed to determine optimal hyperparameter combinations for different datasets. Experimental results on six prevailing datasets demonstrate that MHMOGAT improves GAT accuracy by approximately 2-5% across various datasets with different label rates, indicating its effectiveness. Additionally, MHMOGAT exhibits potential in handling large and complex graphs with low label rates.
Keyword:
Semi-supervised learning
Graph convolutional network
Attention mechanism
Multi-order information
期刊
IF:
3.5
论文数:
7.6K
被引数:
1.7W
机构
引用论文
Análisis espacial de las concentraciones de PM2,5 en Bogotá según los valores de las guías de la calidad del aire de la Organización Mundial de la Salud para enfermedades cardiopulmonares, 2014-2015.png)
Biomédica
OAAI
BiomédicaIF0
OAAI分享收藏Robust semi-supervised support vector machines with Laplace kernel-induced correntropy loss functionsAPPLIED INTELLIGENCE
APPLIED INTELLIGENCEIF3.5
PREAI
分享收藏A gated graph attention network based on dual graph convolution for node embedding基于对偶图卷积的节点嵌入门控图注意力网络APPLIED INTELLIGENCE
APPLIED INTELLIGENCEIF3.5
PREAI
分享收藏
查看更多内容