TripleRE: Knowledge Graph Embeddings via Tripled Relation Vectors
This work addresses knowledge representation learning for knowledge graphs, offering incremental improvements in embedding methods.
The paper tackled the problem of unsatisfactory performance in translation-based knowledge graph embeddings by proposing TripleRE, a method that divides relationship vectors into three parts and uses residual concepts, achieving state-of-the-art results on a large-scale dataset and competitive performance on others.
Translation-based knowledge graph embedding has been one of the most important branches for knowledge representation learning since TransE came out. Although many translation-based approaches have achieved some progress in recent years, the performance was still unsatisfactory. This paper proposes a novel knowledge graph embedding method named TripleRE with two versions. The first version of TripleRE creatively divide the relationship vector into three parts. The second version takes advantage of the concept of residual and achieves better performance. In addition, attempts on using NodePiece to encode entities achieved promising results in reducing the parametric size, and solved the problems of scalability. Experiments show that our approach achieved state-of-the-art performance on the large-scale knowledge graph dataset, and competitive performance on other datasets.