MCMH: Learning Multi-Chain Multi-Hop Rules for Knowledge Graph Reasoning

TL;DR

This paper introduces MCMH, a novel multi-hop reasoning method over knowledge graphs that learns generalized multi-chain rules using a game-theoretical framework, leading to improved inference accuracy.

Contribution

It proposes a new generalized multi-hop rule formulation and a two-step learning approach with a game-theoretical framework for efficient rule learning.

Findings

MCMH outperforms standard single-chain methods in knowledge graph reasoning.

The generalized multi-chain rules improve inference accuracy.

The proposed framework effectively learns complex multi-hop relations.

Abstract

Multi-hop reasoning approaches over knowledge graphs infer a missing relationship between entities with a multi-hop rule, which corresponds to a chain of relationships. We extend existing works to consider a generalized form of multi-hop rules, where each rule is a set of relation chains. To learn such generalized rules efficiently, we propose a two-step approach that first selects a small set of relation chains as a rule and then evaluates the confidence of the target relationship by jointly scoring the selected chains. A game-theoretical framework is proposed to this end to simultaneously optimize the rule selection and prediction steps. Empirical results show that our multi-chain multi-hop (MCMH) rules result in superior results compared to the standard single-chain approaches, justifying both our formulation of generalized rules and the effectiveness of the proposed learning framework.