Quaternion Knowledge Graph Embeddings

By Shuai Zhang, Yi Tay, Lina Yao, Qi Liu. Neural Information Processing Systems, 2019

583 citations recorded by the source corpus.

In this work, we move beyond the traditional complex-valued representations, introducing more expressive hypercomplex representations to model entities and relations for knowledge graph embeddings. More specifically, quaternion embeddings, hypercomplex-valued embeddings with three imaginary components, are utilized to represent entities. Relations are modelled as rotations in the quaternion space. The advantages of the proposed approach are: (1) Latent inter-dependencies (between all components) are aptly captured with Hamilton product, encouraging a more compact interaction between entities and relations; (2) Quaternions enable expressive rotation in four-dimensional space and have more degree of freedom than rotation in complex plane; (3) The proposed framework is a generalization of ComplEx on hypercomplex space while offering better geometrical interpretations, concurrently satisfying the key desiderata of relational representation learning (i.e., modeling symmetry, anti-symmetry and inversion). Experimental results demonstrate that our method achieves state-of-the-art performance on four well-established knowledge graph completion benchmarks.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/f78894c38cfd6404220b2b55b7f0f8c4fecbc34f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Knowledge Graphs: Opportunities and Challenges (2023, background): “One of the biggest problems in this space is fake news (Zhang et al. 2019a).”
HittER: Hierarchical Transformers for Knowledge Graph Embeddings (2020, background): “…in the eu-clidean space (Nickel et al., 2011; Yang et al., 2015), numerous complicated transformations in various spaces have been hence proposed (Trouil-lon et al., 2016; Ebisu and Ichise, 2018; Sun et al., 2018; Zhang et al., 2019a; Chami et al., 2020; Tang et al., 2020; Chao et al., 2021).”
Low-Dimensional Hyperbolic Knowledge Graph Embeddings (2020, methodology): “The recent QuatE model (Zhang et al., 2019) learns KG embeddings using quaternions.”
Exploiting Structured Knowledge in Text via Graph-Guided Representation Learning (2020, methodology): “, 2019c), (Zhang et al., 2019a), (Yao et al.”
Knowledge Graph Completion: A Review (2020, methodology): “The model QuatE [71] uses two rotating planes in the complex space to preserve the symmetry/antisymmetric, ﬂip, and combination relations, and models the relations as rotations in a hyper-complex space, thus unifying ComplEx and RotatE.”
Lightweight and Efficient Neural Natural Language Processing with Quaternion Networks (2019, unknown): “(Zhang et al., 2019b) proposed Quaternion representations for collaborative ﬁlter-ing.”
Relational Memory-Augmented Language Models (2022, background): “However, existing knowledge bases suffer from missing relations and often do not contain relation triples related to observed contexts in a target corpus, even though research on knowledge base completion has resulted in significant advances (Bordes et al., 2013; Trouillon et al., 2016; Zhang et al.”
A Survey on Knowledge Graphs: Representation, Acquisition, and Applications (2020, methodology): “QuatE [24] extends the complex-valued space into hypercomplex h, t, r ∈ H by a quaternion Q = a + bi + cj + dk with three imaginary components, where the quaternion inner product, i.”
A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multi-Modal (2022, background): “QuatE[52]empowers the semantic matching between head and tail based on relational rotation quaternion representations.”