On the representation and embedding of knowledge bases beyond binary relations

TL;DR

This paper introduces a new framework for representing and embedding multi-fold relations in knowledge bases, improving over traditional binary relation models and achieving state-of-the-art results.

Contribution

It proposes a canonical representation for multi-fold relations and generalizes existing models like TransH to better capture structural information.

Findings

m-TransH outperforms TransH significantly

The new framework preserves more structural information

Achieves state-of-the-art results on FB15K dataset

Abstract

The models developed to date for knowledge base embedding are all based on the assumption that the relations contained in knowledge bases are binary. For the training and testing of these embedding models, multi-fold (or n-ary) relational data are converted to triples (e.g., in FB15K dataset) and interpreted as instances of binary relations. This paper presents a canonical representation of knowledge bases containing multi-fold relations. We show that the existing embedding models on the popular FB15K datasets correspond to a sub-optimal modelling framework, resulting in a loss of structural information. We advocate a novel modelling framework, which models multi-fold relations directly using this canonical representation. Using this framework, the existing TransH model is generalized to a new model, m-TransH. We demonstrate experimentally that m-TransH outperforms TransH by a large margin, thereby establishing a new state of the art.