arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.03147cs.LG
收藏

图神经网络的联合注意力学习

Learning Conjoint Attentions for Graph Neural Nets

Nanyang Technological University(南洋理工大学);Agency for Science, Technology and Research(新加坡科技研究局)

Tiantian He, Yew-Soon Ong, Lu Bai

摘要:

在本文中,我们提出了联合注意力(CA),一类新的学习参加图神经网络(GNNs)的战略。除了考虑在GNN内传播的逐层节点特征外,CA还可以在计算注意力分数时结合各种结构干预,例如节点簇嵌入和可以在GNN外部学习的高阶结构相关性。因此,被联合标准视为重要的节点特征更有可能在GNN中传播。鉴于新的联合注意力策略,我们提出了图联合注意力网络(CAT),它可以学习嵌入有联合注意力所认为的重要潜在特征的表示。此外,我们从理论上验证了CAT的辨别能力。利用所提出的联合注意力策略的CAT已经在成熟的基准数据集中进行了广泛的测试,并与最先进的基线进行了全面的比较。所获得的显着性能证明了所提出的联合注意力的有效性。

英文摘要:

In this paper, we present Conjoint Attentions (CAs), a class of novel learning-to-attend strategies for graph neural networks (GNNs). Besides considering the layer-wise node features propagated within the GNN, CAs can additionally incorporate various structural interventions, such as node cluster embedding, and higher-order structural correlations that can be learned outside of GNN, when computing attention scores. The node features that are regarded as significant by the conjoint criteria are therefore more likely to be propagated in the GNN. Given the novel Conjoint Attention strategies, we then propose Graph conjoint attention networks (CATs) that can learn representations embedded with significant latent features deemed by the Conjoint Attentions. Besides, we theoretically validate the discriminative capacity of CATs. CATs utilizing the proposed Conjoint Attention strategies have been extensively tested in well-established benchmarking datasets and comprehensively compared with state-of-the-art baselines. The obtained notable performance demonstrates the effectiveness of the proposed Conjoint Attentions.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑