HyperKG: Hyperbolic Knowledge Graph Embeddings for Knowledge Base Completion
This work addresses knowledge base completion for AI applications, offering an incremental improvement by applying hyperbolic geometry to an existing model family.
The authors tackled the problem of knowledge base completion by proposing HyperKG, a translational model that uses hyperbolic embeddings to better capture topological properties, narrowing the performance gap with bilinear models on link prediction datasets.
Learning embeddings of entities and relations existing in knowledge bases allows the discovery of hidden patterns in data. In this work, we examine the geometrical space's contribution to the task of knowledge base completion. We focus on the family of translational models, whose performance has been lagging, and propose a model, dubbed HyperKG, which exploits the hyperbolic space in order to better reflect the topological properties of knowledge bases. We investigate the type of regularities that our model can capture and we show that it is a prominent candidate for effectively representing a subset of Datalog rules. We empirically show, using a variety of link prediction datasets, that hyperbolic space allows to narrow down significantly the performance gap between translational and bilinear models.