RotateQVS: Representing Temporal Information as Rotations in Quaternion Vector Space for Temporal Knowledge Graph Completion

Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics Volume 1: Long Papers, pages 5843 - 5857 May 22-27, 2022 c⃝2022 Association for Computational Linguistics RotateQVS: Representing Temporal Information as Rotations in Quaternion Vector Space for Temporal Knowledge Graph Completion Kai Chen1, Ye Wang1,2∗, Yitong Li3, Aiping Li1 1National University of Defense Technology, Changsha, China 2Pengcheng Laboratory, Shenzhen, China 3Huawei Technologies Co., Ltd. {chenkai_,ye.wang,liaiping}@nudt.edu.cn liyitong3@huawei.com Abstract Temporal factors are tied to the growth of facts in realistic applications, such as the progress of diseases and the development of political situ- ation, therefore, research on Temporal Knowl- edge Graph (TKG) attracks much attention. In TKG, relation patterns inherent with temporal- ity are required to be studied for representation learning and reasoning across temporal facts. However, existing methods can hardly model temporal relation patterns, nor can capture the intrinsic connections between relations when evolving over time, lacking of interpretability. In this paper, we propose a novel temporal mod- eling method which represents temporal enti- ties as Rotations in Quaternion Vector Space (RotateQVS) and relations as complex vectors in Hamilton’s quaternion space. We demon- strate our method can model key patterns of re- lations in TKG, such as symmetry, asymmetry, inverse, and can further capture time-evolved relations by theory. Empirically, we show that our method can boost the performance of link prediction tasks over four temporal knowledge graph benchmarks. 1 Introduction Knowledge Graphs (KGs) have been widely adopted to represent informative knowledge or facts in real-world applications (Bollacker et al., 2008; Miller, 1995; Suchanek et al., 2007). How- ever, as known facts are usually sparse, KGs are far from completeness. Thus, Knowledge Graph Completion (KGC) methods are proposed to pre- dict missing facts, i.e. links between entities (Bor- des et al., 2013; Yang et al., 2015; Dettmers et al., 2018; Chen et al., 2021b). Furthermore, in real world, many facts are bonded with a particular time by nature. For example, Barack Obama is the president of USA is only valid for the time pe- riod 2009 - 2017. To model such time-sensitive facts, Temporal Knowledge Graphs (TKGs) have ∗ Corresponding author. recently drawn growing attention from both aca- demic and industrial communities (Lautenschlager et al., 2015; Leetaru and Schrodt, 2013). TKG Embedding (TKGE) methods (Jiang et al., 2016; Dasgupta et al., 2018; Jin et al., 2020; Sadeghian et al., 2021) were proposed to repre- sent entities and relations with temporal features in TKGs (Lautenschlager et al., 2015; Leetaru and Schrodt, 2013). But how to present them with temporal interpretability remains a challenge for state-of-the-art TKGE models. Further, it is cru- cial for TKG Completion (TKGC) to leverage the learned temporal information. Previous static KGC works (Sun et al., 2020; Schlichtkrull et al., 2018; Gao et al., 2020) learn explainable embeddings of various relation patterns, so that symmetric pattern (e.g. “co-author”), asymmetric pattern (e.g. “affili- ation”), inverse pattern (e.g. “buyer” vs. “seller”) and complex composition pattern (e.g. “father’s wife (mother)” vs. “wife’s father (father in law)”) can be captured in static …

Content truncated to 500 words