Logical Reasoning with Relation Network for Inductive Knowledge Graph Completion

TL;DR

This paper introduces NORAN, a relation network that captures relation patterns to improve inductive knowledge graph completion, especially addressing data sparsity and cold-start issues, outperforming existing methods.

Contribution

The paper proposes NORAN, a novel relation network that models relation correlations for inductive KGC, enhancing reasoning over new entities with limited local information.

Findings

NORAN significantly outperforms state-of-the-art methods on five benchmarks.

Relation-centered modeling effectively captures logical evidence for inductive reasoning.

NORAN addresses data sparsity and cold-start problems in inductive KGC.

Abstract

Inductive knowledge graph completion (KGC) aims to infer the missing relation for a set of newly-coming entities that never appeared in the training set. Such a setting is more in line with reality, as real-world KGs are constantly evolving and introducing new knowledge. Recent studies have shown promising results using message passing over subgraphs to embed newly-coming entities for inductive KGC. However, the inductive capability of these methods is usually limited by two key issues. (i) KGC always suffers from data sparsity, and the situation is even exacerbated in inductive KGC where new entities often have few or no connections to the original KG. (ii) Cold-start problem. It is over coarse-grained for accurate KG reasoning to generate representations for new entities by gathering the local information from few neighbors. To this end, we propose a novel iNfOmax RelAtion Network, namely NORAN, for inductive KG completion. It aims to mine latent relation patterns for inductive KG completion. Specifically, by centering on relations, NORAN provides a hyper view towards KG modeling, where the correlations between relations can be naturally captured as entity-independent logical evidence to conduct inductive KGC. Extensive experiment results on five benchmarks show that our framework substantially outperforms the state-of-the-art KGC methods.