Simple but Effective Compound Geometric Operations for Temporal Knowledge Graph Completion

TL;DR

This paper introduces TCompoundE, a novel method for temporal knowledge graph completion that employs two geometric operations to better capture complex temporal dynamics, significantly outperforming existing models.

Contribution

The paper proposes TCompoundE, a new approach with dual geometric operations tailored for temporal knowledge graphs, enhancing pattern encoding capabilities.

Findings

TCompoundE outperforms existing models in experiments.

Mathematical proofs demonstrate encoding of various relation patterns.

The method effectively captures complex temporal dynamics.

Abstract

Temporal knowledge graph completion aims to infer the missing facts in temporal knowledge graphs. Current approaches usually embed factual knowledge into continuous vector space and apply geometric operations to learn potential patterns in temporal knowledge graphs. However, these methods only adopt a single operation, which may have limitations in capturing the complex temporal dynamics present in temporal knowledge graphs. Therefore, we propose a simple but effective method, i.e. TCompoundE, which is specially designed with two geometric operations, including time-specific and relation-specific operations. We provide mathematical proofs to demonstrate the ability of TCompoundE to encode various relation patterns. Experimental results show that our proposed model significantly outperforms existing temporal knowledge graph embedding models. Our code is available at https://github.com/nk-ruiying/TCompoundE.