Group Representation Theory for Knowledge Graph Embedding
This work provides a theoretical foundation for knowledge graph embedding methods, which is incremental as it builds on and generalizes existing approaches like RotatE.
The paper tackles the problem of modeling relations in knowledge graph embedding by connecting existing methods to group actions and using Schur's lemma to show that the RotatE method can model relations from any finite Abelian group, achieving a theoretical extension without new empirical results.
Knowledge graph embedding has recently become a popular way to model relations and infer missing links. In this paper, we present a group theoretical perspective of knowledge graph embedding, connecting previous methods with different group actions. Furthermore, by utilizing Schur's lemma from group representation theory, we show that the state of the art embedding method RotatE can model relations from any finite Abelian group.