Search to Pass Messages for Temporal Knowledge Graph Completion

TL;DR

This paper introduces a neural architecture search framework to automatically design message passing architectures for temporal knowledge graph completion, achieving state-of-the-art results and revealing diverse properties of TKGs.

Contribution

It proposes a generalized NAS framework for TKGs, enabling data-specific architecture design that captures diverse topological and temporal properties.

Findings

Achieves state-of-the-art performance on three benchmark datasets.

Automatically discovers architectures that reveal diverse TKG properties.

Efficient search with less computational cost.

Abstract

Completing missing facts is a fundamental task for temporal knowledge graphs (TKGs). Recently, graph neural network (GNN) based methods, which can simultaneously explore topological and temporal information, have become the state-of-the-art (SOTA) to complete TKGs. However, these studies are based on hand-designed architectures and fail to explore the diverse topological and temporal properties of TKG. To address this issue, we propose to use neural architecture search (NAS) to design data-specific message passing architecture for TKG completion. In particular, we develop a generalized framework to explore topological and temporal information in TKGs. Based on this framework, we design an expressive search space to fully capture various properties of different TKGs. Meanwhile, we adopt a search algorithm, which trains a supernet structure by sampling single path for efficient search with less cost. We further conduct extensive experiments on three benchmark datasets. The results show that the searched architectures by our method achieve the SOTA performances. Besides, the searched models can also implicitly reveal diverse properties in different TKGs. Our code is released in https://github.com/striderdu/SPA.