Filters








373,946 Hits in 2.5 sec

Graph Attention Networks [article]

Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, Yoshua Bengio
2018 arXiv   pre-print
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  ...  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.  ...  CONCLUSIONS We have presented graph attention networks (GATs), novel convolution-style neural networks that operate on graph-structured data, leveraging masked self-attentional layers.  ... 
arXiv:1710.10903v3 fatcat:6xbvqpuxsjfuhfo7vqcdjmefgq

How Attentive are Graph Attention Networks? [article]

Shaked Brody, Uri Alon, Eran Yahav
2022 arXiv   pre-print
Graph Attention Networks (GATs) are one of the most popular GNN architectures and are considered as the state-of-the-art architecture for representation learning with graphs.  ...  Because GATs use a static attention mechanism, there are simple graph problems that GAT cannot express: in a controlled problem, we show that static attention hinders GAT from even fitting the training  ...  GNN variants -Graph Attention Network (GAT).  ... 
arXiv:2105.14491v3 fatcat:pp4xs7d2bffo7jb2p2zn626lya

Graph Ordering Attention Networks [article]

Michail Chatzianastasis, Johannes F. Lutzeyer, George Dasoulas, Michalis Vazirgiannis
2022 arXiv   pre-print
Graph Neural Networks (GNNs) have been successfully used in many problems involving graph-structured data, achieving state-of-the-art performance.  ...  This is achieved by learning local node orderings via an attention mechanism and processing the ordered representations using a recurrent neural network aggregator.  ...  Graph Neural Networks.  ... 
arXiv:2204.05351v2 fatcat:ptxpj6ghh5aw5kcl6vifsf5ghu

Ensemble Graph Attention Networks

Nan Wu, Chaofan Wang
2022 Transactions on Machine Learning and Artificial Intelligence  
We then propose two ensemble graph neural network models – Ensemble-GAT and Ensemble-HetGAT by applying the ensemble strategy to the graph attention network (GAT), and a heterogeneous graph attention network  ...  Graph neural networks have demonstrated its success in many applications on graph-structured data.  ...  Ensemble Graph Attention Networks.  ... 
doi:10.14738/tmlai.103.12399 fatcat:qivrmha2gjhwrft4ejhfomk5ti

Hyperbolic Graph Attention Network [article]

Yiding Zhang, Xiao Wang, Xunqiang Jiang, Chuan Shi, Yanfang Ye
2019 arXiv   pre-print
Graph neural network (GNN) has shown superior performance in dealing with graphs, which has attracted considerable research attention recently.  ...  The comprehensive experimental results on four real-world datasets demonstrate the performance of our proposed hyperbolic graph attention network model, by comparisons with other state-of-the-art baseline  ...  Acknowledgements Hyperbolic representation learning has attracted considerable research attention recently.  ... 
arXiv:1912.03046v1 fatcat:ag2elsxc4bdyfakvf72juawz2i

Heterogeneous Graph Attention Network [article]

Xiao Wang, Houye Ji, Chuan Shi, Bai Wang, Peng Cui, P. Yu, Yanfang Ye
2021 arXiv   pre-print
In this paper, we first propose a novel heterogeneous graph neural network based on the hierarchical attention, including node-level and semantic-level attentions.  ...  The heterogeneity and rich semantic information bring great challenges for designing a graph neural network for heterogeneous graph.  ...  Graph Attention Network (GAT) [35] , a novel convolution-style graph neural network, leverages attention mechanism for the homogeneous graph which includes only one type of nodes or links.  ... 
arXiv:1903.07293v2 fatcat:wrwdajlnm5bahhh26ngd5z7pfe

Sparse Graph Attention Networks [article]

Yang Ye, Shihao Ji
2021 arXiv   pre-print
Among the variants of GNNs, Graph Attention Networks (GATs) learn to assign dense attention coefficients over all neighbors of a node for feature aggregation, and improve the performance of many graph  ...  In this paper, we propose Sparse Graph Attention Networks (SGATs) that learn sparse attention coefficients under an L_0-norm regularization, and the learned sparse attentions are then used for all GNN  ...  CONCLUSION In this paper we propose sparse graph attention networks (SGATs) that integrate a sparse attention mechanism into graph attention networks (GATs) via an L 0 -norm regularization on the number  ... 
arXiv:1912.00552v2 fatcat:vqvo7aty6beyphlgtgi5qsg7fq

Signed Graph Attention Networks [article]

Junjie Huang, Huawei Shen, Liang Hou, Xueqi Cheng
2019 arXiv   pre-print
In this paper, we propose Signed Graph Attention Networks (SiGATs), generalizing GAT to signed networks.  ...  As a representative implementation of GNNs, Graph Attention Networks (GATs) are successfully applied in a variety of tasks on real datasets.  ...  Signed Graph Attention Networks GAT [23] introduces attention mechanism into the graph. It uses a weight matrix to characterize the different effects of different nodes on the target node.  ... 
arXiv:1906.10958v3 fatcat:jx6eeecytfgm3d4d22rap2gh4a

Relational Graph Attention Networks [article]

Dan Busbridge, Dane Sherburn, Pietro Cavallo, Nils Y. Hammerla
2019 arXiv   pre-print
We investigate Relational Graph Attention Networks, a class of models that extends non-relational graph attention mechanisms to incorporate relational information, opening up these methods to a wider variety  ...  To provide a meaningful comparison, we retrain Relational Graph Convolutional Networks, the spectral counterpart of Relational Graph Attention Networks, and evaluate them under the same conditions.  ...  Conclusion We have investigated a class of models we call Relational Graph Attention Networks (RGATs).  ... 
arXiv:1904.05811v1 fatcat:blea66xmsvflfffhbpmffpzwyu

Universal Graph Transformer Self-Attention Networks [article]

Dai Quoc Nguyen and Tu Dinh Nguyen and Dinh Phung
2022 arXiv   pre-print
We introduce a transformer-based GNN model, named UGformer, to learn graph representations.  ...  Experimental results demonstrate that the first UGformer variant achieves state-of-the-art accuracies on benchmark datasets for graph classification in both inductive setting and unsupervised transductive  ...  It is worth noting that all links/interactions among all positions in the self-attention layer build up a complete network.  ... 
arXiv:1909.11855v13 fatcat:eqc3zguszvdhden5ycwdz75rze

Attention Guided Graph Convolutional Networks for Relation Extraction [article]

Zhijiang Guo and Yan Zhang and Wei Lu
2020 arXiv   pre-print
In this work, we propose Attention Guided Graph Convolutional Networks (AGGCNs), a novel model which directly takes full dependency trees as inputs.  ...  In this paper, we propose the novel Attention Guided Graph Convolutional Networks (AGGCNs), which operate directly on the full tree.  ...  Graph Convolutional Networks. Early efforts that attempt to extend neural networks to deal with arbitrary structured graphs are introduced by Gori et al. (2005); Bruna (2014) .  ... 
arXiv:1906.07510v8 fatcat:duxwaeueorb5dkhanhg2vlv32m

SGAT: Simplicial Graph Attention Network [article]

See Hian Lee, Feng Ji, Wee Peng Tay
2022 arXiv   pre-print
In this paper, we present Simplicial Graph Attention Network (SGAT), a simplicial complex approach to represent such high-order interactions by placing features from non-target nodes on the simplices.  ...  To learn such complex semantics, many graph neural network approaches for heterogeneous graphs use metapaths to capture multi-hop interactions between nodes.  ...  Recently, graph neural network (GNN), which learns tasks utilizing graph-structured data, has attracted much attention.  ... 
arXiv:2207.11761v1 fatcat:lonsnt54jjeg7gya6wyliwjuvu

Personalized PageRank Graph Attention Networks [article]

Julie Choi
2022 arXiv   pre-print
In this work, we incorporate the limit distribution of Personalized PageRank (PPR) into graph attention networks (GATs) to reflect the larger neighbor information without introducing over-smoothing.  ...  There has been a rising interest in graph neural networks (GNNs) for representation learning over the past few years.  ...  Graph Attention Networks Recently, attention networks have achieved state-of-the-art results in many tasks [14] .  ... 
arXiv:2205.14259v1 fatcat:xyccmktkobf4rdvtipxpsbog6m

Dual-Attention Graph Convolutional Network [article]

Xueya Zhang and Tong Zhang and Wenting Zhao and Zhen Cui and Jian Yang
2019 arXiv   pre-print
Graph convolutional networks (GCNs) have shown the powerful ability in text structure representation and effectively facilitate the task of text classification.  ...  In this paper, we propose a dual-attention GCN to model the structural information of various texts as well as tackle the graph-invariant problem through embedding two types of attention mechanisms, i.e  ...  Graph Convolutional Network. In recent years, graph convolutional networks (GCNs) gain more attention because of some unique advantages.  ... 
arXiv:1911.12486v1 fatcat:a5x4f3ypknhdrb7esmdgndpwzy

Context-Aware Graph Attention Networks [article]

Bo Jiang, Leiling Wang, Jin Tang, Bin Luo
2019 arXiv   pre-print
In this paper, we propose a novel unified GNN model, named Context-aware Adaptive Graph Attention Network (CaGAT).  ...  Graph Neural Networks (GNNs) have been widely studied for graph data representation and learning.  ...  Conclusion In this paper, we propose a novel spatial graph neural network, named Context-aware Adaptive Graph Attention Network (CaGAT).  ... 
arXiv:1910.01736v1 fatcat:4g6t2tprananfh7k4dfjrxh7ze
« Previous Showing results 1 — 15 out of 373,946 results