Subgraph Neighboring Relations Infomax for Inductive Link Prediction on Knowledge Graphs
This addresses the challenge of handling unseen entities in knowledge graphs, which is crucial for real-world applications, though it is an incremental advance over existing subgraph-based methods.
The paper tackles the problem of inductive link prediction on knowledge graphs, where unseen entities lack embeddings, by proposing SNRI which exploits complete neighboring relations and uses mutual information maximization, achieving state-of-the-art performance with significant improvements.
Inductive link prediction for knowledge graph aims at predicting missing links between unseen entities, those not shown in training stage. Most previous works learn entity-specific embeddings of entities, which cannot handle unseen entities. Recent several methods utilize enclosing subgraph to obtain inductive ability. However, all these works only consider the enclosing part of subgraph without complete neighboring relations, which leads to the issue that partial neighboring relations are neglected, and sparse subgraphs are hard to be handled. To address that, we propose Subgraph Neighboring Relations Infomax, SNRI, which sufficiently exploits complete neighboring relations from two aspects: neighboring relational feature for node feature and neighboring relational path for sparse subgraph. To further model neighboring relations in a global way, we innovatively apply mutual information (MI) maximization for knowledge graph. Experiments show that SNRI outperforms existing state-of-art methods by a large margin on inductive link prediction task, and verify the effectiveness of exploring complete neighboring relations in a global way to characterize node features and reason on sparse subgraphs.