Sharing Parameter by Conjugation for Knowledge Graph Embeddings in Complex Space

TL;DR

This paper introduces a parameter-sharing technique using conjugate parameters for complex-valued knowledge graph embeddings, significantly reducing memory usage while maintaining performance and training efficiency across multiple models and datasets.

Contribution

It proposes a novel conjugate parameter-sharing method that enhances memory efficiency and training speed in complex space KGE models without sacrificing accuracy.

Findings

Memory usage reduced by 2x in relation embeddings

Achieved comparable performance to state-of-the-art models

Demonstrated generalizability on multiple datasets and models

Abstract

A Knowledge Graph (KG) is the directed graphical representation of entities and relations in the real world. KG can be applied in diverse Natural Language Processing (NLP) tasks where knowledge is required. The need to scale up and complete KG automatically yields Knowledge Graph Embedding (KGE), a shallow machine learning model that is suffering from memory and training time consumption issues. To mitigate the computational load, we propose a parameter-sharing method, i.e., using conjugate parameters for complex numbers employed in KGE models. Our method improves memory efficiency by 2x in relation embedding while achieving comparable performance to the state-of-the-art non-conjugate models, with faster, or at least comparable, training time. We demonstrated the generalizability of our method on two best-performing KGE models $5^{\bigstar}\mathrm{E}$ and $\mathrm{ComplEx}$ on five benchmark datasets.