Published October 4, 2021 | Version v1
Conference paper Open

Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations

  • 1. University College London, Facebook AI Research London
  • 2. University College London

Description

Learning good representations on multi-relational graphs is essential to knowledge base completion (KBC). In this paper, we propose a new self-supervised training objective for multi-relational graph representation learning, via simply incorporating relation prediction into the commonly used 1vsAll objective. The new training objective contains not only terms for predicting the subject and object of a given triple, but also a term for predicting the relation type. We analyse how this new objective impacts multi-relational learning in KBC: experiments on a variety of datasets and models show that relation prediction can significantly improve entity ranking, the most widely used evaluation task for KBC, yielding a 6.1% increase in MRR and 9.9% increase in Hits@1 on FB15k-237 as well as a 3.1% increase in MRR and 3.4% in Hits@1 on Aristo-v4. Moreover, we observe that the proposed objective is especially effective on highly multi-relational datasets, i.e. datasets with a large number of predicates, and generates better representations when larger embedding sizes are used.

Notes

AKBC 2021. Code available at https://github.com/facebookresearch/ssl-relation-prediction.

Files

2110.02834.pdf

Files (2.9 MB)

Name Size Download all
md5:162e3827e444138a39242c3ca4957be0
985.3 kB Preview Download
md5:4e0e9be2808b9b43194494ba61f8f97b
1.9 MB Preview Download