arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2109.05922cs.AIcs.CLcs.LG
收藏

r-GAT:多关系图的关系图注意力网络

r-GAT: Relational Graph Attention Network for Multi-Relational Graphs

Peking University(北京大学)

Meiqi Chen, Yuan Zhang, Xiaoyu Kou, Yuntao Li, Yan Zhang

摘要:

图注意力网络(GAT)专注于建模简单的无向和单一的关系图数据。这限制了它处理更一般和复杂的多关系图的能力,这些图包含具有不同标签的有向链接的实体(例如,知识图谱)。因此,直接将GAT应用于多关系图会导致次优解决方案。为了解决这个问题,我们提出了r-GAT,一个关系图注意力网络来学习多通道实体表示。具体地,每个通道对应于实体的潜在语义方面。这使我们能够使用关系特征来聚合当前方面的邻域信息。我们进一步提出了一个查询感知的注意力机制,为后续的任务选择有用的方面。在链接预测和实体分类任务上的大量实验表明,我们的r-GAT可以有效地建模多关系图。此外,我们通过案例研究表明我们的方法的可解释性。

英文摘要:

Graph Attention Network (GAT) focuses on modelling simple undirected and single relational graph data only. This limits its ability to deal with more general and complex multi-relational graphs that contain entities with directed links of different labels (e.g., knowledge graphs). Therefore, directly applying GAT on multi-relational graphs leads to sub-optimal solutions. To tackle this issue, we propose r-GAT, a relational graph attention network to learn multi-channel entity representations. Specifically, each channel corresponds to a latent semantic aspect of an entity. This enables us to aggregate neighborhood information for the current aspect using relation features. We further propose a query-aware attention mechanism for subsequent tasks to select useful aspects. Extensive experiments on link prediction and entity classification tasks show that our r-GAT can model multi-relational graphs effectively. Also, we show the interpretability of our approach by case study.

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。

↑