Distance-Based Propagation for Efficient Knowledge Graph Reasoning

TL;DR

This paper introduces TAGNet, a novel method for knowledge graph completion that efficiently propagates information by limiting path aggregation, significantly reducing message passing while maintaining competitive accuracy.

Contribution

The paper proposes TAGNet, a new approach that addresses efficiency and quality limitations in existing path-aggregation methods for knowledge graph reasoning.

Findings

Reduces propagated messages by up to 90%.

Maintains competitive performance across multiple datasets.

Complexity is independent of the number of layers.

Abstract

Knowledge graph completion (KGC) aims to predict unseen edges in knowledge graphs (KGs), resulting in the discovery of new facts. A new class of methods have been proposed to tackle this problem by aggregating path information. These methods have shown tremendous ability in the task of KGC. However they are plagued by efficiency issues. Though there are a few recent attempts to address this through learnable path pruning, they often sacrifice the performance to gain efficiency. In this work, we identify two intrinsic limitations of these methods that affect the efficiency and representation quality. To address the limitations, we introduce a new method, TAGNet, which is able to efficiently propagate information. This is achieved by only aggregating paths in a fixed window for each source-target pair. We demonstrate that the complexity of TAGNet is independent of the number of layers. Extensive experiments demonstrate that TAGNet can cut down on the number of propagated messages by as much as 90% while achieving competitive performance on multiple KG datasets. The code is available at https://github.com/HarryShomer/TAGNet.