Simple Rule Injection for ComplEx Embeddings

TL;DR

This paper introduces InjEx, a simple and scalable method to inject multiple types of logical rules into knowledge graph embeddings, improving inference performance in KGC and FKGC tasks.

Contribution

InjEx is a novel mechanism that injects multiple rule types via simple constraints, theoretically proven and empirically shown to enhance knowledge graph completion tasks.

Findings

InjEx outperforms baseline models in KGC and FKGC tasks.

InjEx maintains scalability and efficiency.

Theoretical proof of rule injection capability.

Abstract

Recent works in neural knowledge graph inference attempt to combine logic rules with knowledge graph embeddings to benefit from prior knowledge. However, they usually cannot avoid rule grounding, and injecting a diverse set of rules has still not been thoroughly explored. In this work, we propose InjEx, a mechanism to inject multiple types of rules through simple constraints, which capture definite Horn rules. To start, we theoretically prove that InjEx can inject such rules. Next, to demonstrate that InjEx infuses interpretable prior knowledge into the embedding space, we evaluate InjEx on both the knowledge graph completion (KGC) and few-shot knowledge graph completion (FKGC) settings. Our experimental results reveal that InjEx outperforms both baseline KGC models as well as specialized few-shot models while maintaining its scalability and efficiency.