Inductive Relation Prediction by Subgraph Reasoning

Abstract

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.

References in corpus (8)

• Inductive Representation Learning on Large Graphs
• Geometric deep learning: going beyond Euclidean data
• A Review of Relational Machine Learning for Knowledge Graphs
• RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space
• Knowledge Transfer for Out-of-Knowledge-Base Entities: A Graph Neural Network Approach
• TensorLog: A Differentiable Deductive Database
• DRUM: End-To-End Differentiable Rule Mining On Knowledge Graphs
• Knowledge Graph Completion via Complex Tensor Factorization

Cited by in corpus (10)

• Construction of Knowledge Graphs: State and Challenges
• xERTE: Explainable Reasoning on Temporal Knowledge Graphs for Forecasting Future Links
• BioBLP: A Modular Framework for Learning on Multimodal Biomedical Knowledge Graphs
• Logical Reasoning with Relation Network for Inductive Knowledge Graph Completion
• Data Considerations in Graph Representation Learning for Supply Chain Networks
• Position-Aware Subgraph Neural Networks with Data-Efficient Learning
• Learning First-Order Rules with Relational Path Contrast for Inductive Relation Reasoning
• Neural, Symbolic and Neural-Symbolic Reasoning on Knowledge Graphs
• OntoZSL: Ontology-enhanced Zero-shot Learning
• Node Attribute Completion in Knowledge Graphs with Multi-Relational Propagation