📰 Relation Prediction as A Self-Supervised Learning Objective

Abstract

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.

Useful Links

• Check the full paper and slides at AKBC 2021.
• Check the code at facebookresearch/ssl-relation-prediction and feel free to ping me if you need any help with the code.
• Check how our method performs compared with other KBC methods at paperswithcode.

Citation

If you find our work helpful, please cite us

@inproceedings{
chen2021relation,
title={Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations},
author={Yihong Chen and Pasquale Minervini and Sebastian Riedel and Pontus Stenetorp},
booktitle={3rd Conference on Automated Knowledge Base Construction},
year={2021},
url={https://openreview.net/forum?id=Qa3uS3H7-Le}
}