Inductively Representing Out-of-Knowledge-Graph Entities by Optimal Estimation Under Translational Assumptions
This addresses the challenge of efficiently handling evolving knowledge graphs with new entities for applications like knowledge graph completion, though it is incremental as it builds on existing translational assumptions.
The paper tackles the problem of representing out-of-knowledge-graph (OOKG) entities in knowledge graph completion without costly retraining, proposing a method that inductively estimates these entities using pretrained embeddings under translational assumptions, and it outperforms state-of-the-art methods with higher efficiency on two tasks.
Conventional Knowledge Graph Completion (KGC) assumes that all test entities appear during training. However, in real-world scenarios, Knowledge Graphs (KG) evolve fast with out-of-knowledge-graph (OOKG) entities added frequently, and we need to represent these entities efficiently. Most existing Knowledge Graph Embedding (KGE) methods cannot represent OOKG entities without costly retraining on the whole KG. To enhance efficiency, we propose a simple and effective method that inductively represents OOKG entities by their optimal estimation under translational assumptions. Given pretrained embeddings of the in-knowledge-graph (IKG) entities, our method needs no additional learning. Experimental results show that our method outperforms the state-of-the-art methods with higher efficiency on two KGC tasks with OOKG entities.