Inductive Relation Prediction by Subgraph Reasoning

By Komal K. Teru, E. Denis, William L. Hamilton. International Conference on Machine Learning, 2019

524 citations recorded by the source corpus.

The dominant paradigm for relation prediction in knowledge graphs involves learning and operating on latent representations (i.e., embeddings) of entities and relations. However, these embedding-based methods do not explicitly capture the compositional logical rules underlying the knowledge graph, and they are limited to the transductive setting, where the full set of entities must be known during training. Here, we propose a graph neural network based relation prediction framework, GraIL, that reasons over local subgraph structures and has a strong inductive bias to learn entity-independent relational semantics. Unlike embedding-based models, GraIL is naturally inductive and can generalize to unseen entities and graphs after training. We provide theoretical proof and strong empirical evidence that GraIL can represent a useful subset of first-order logic and show that GraIL outperforms existing rule-induction baselines in the inductive setting. We also demonstrate significant gains obtained by ensembling GraIL with various knowledge graph embedding methods in the transductive setting, highlighting the complementary inductive bias of our method.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/c1d9583bc15f622773305fe033fccb67e15c8a49

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jianfeng Gao
2. Xiaodong Liu
3. Ruofei Zhang
4. Jian Jiao
5. Yangfeng Ji
6. Jian Tang
7. Zhaocheng Zhu
8. Zuobai Zhang

In Computer Vision and Pattern Recognition

1. Jianfeng Gao
2. Xiaodong Liu
3. Ruofei Zhang
4. Jian Jiao
5. Xinwang Liu
6. Jian Tang
7. Yue Liu
8. Siwei Wang

Citations in the corpus

• Neural, symbolic and neural-symbolic reasoning on knowledge graphs (2021, background): “Illustration of a knowledge graph (Teru et al., 2019) and examples of two reasoning tasks.”
• HittER: Hierarchical Transformers for Knowledge Graph Embeddings (2020, methodology): “We train and evaluate our proposed method on two standard benchmark datasets FB15K-237 (Toutanova and Chen, 2015) and WN18RR (Dettmers et al., 2018) for link prediction, following the standard train/test split.”
• Neural Bellman-Ford Networks: A General Graph Neural Network Framework for Link Prediction (2021, methodology): “Results of compared methods are taken from [55].”
• A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multi-Modal (2022, methodology): “The prototype GraIL [9], as the landmark inductive GNN-based model, first leverages RGCN to perform the reasoning based on the local enclosingsubgraph.”