知识图谱系列——RGHAT模型学习

Relational Graph Neural Network with Hierarchical Attention for Knowledge Graph Completion

AAAI 2020

---

Reference Format:

GB/T 7714
Zhang Z, Zhuang F, Zhu H, et al. Relational graph neural network with hierarchical attention for knowledge graph completion[C]//Proceedings of the AAAI Conference on Artificial Intelligence. 2020, 34(05): 9612-9619.

---

Datasets:

FB15k、WN18、FB15k-237、WN18RR

---

Problem:

Most existing KGC models treat the triples in KGs independently without leveraging the inherent and valuable information from the local neighborhood surrounding an entity. (Relations and neighborhood entities related to one entity)

---

Methods:

Two-level attention mechanism.
Relation-level attention <=== Different relations have different weights for indicating an entity.
Entity-level attention <=== Weights of neighboring entities under same relation may also be different.

---

---

Points to learn:

1. 本质上相当于利用图神经网络优化实体的嵌入式表示，multi-hop的使用增强了不同实体之间的交互。
2. Decoder的使用大大增加了模型的通用性。

---

该文献未开源