Meta-Learning Based Knowledge Extrapolation for Knowledge Graphs in the Federated Setting

TL;DR

This paper introduces a meta-learning approach using graph neural networks to effectively embed unseen entities and relations in emerging knowledge graphs within a federated setting, outperforming existing methods.

Contribution

It proposes a novel meta-learning framework that constructs features for unseen KG components, enabling better knowledge extrapolation in federated environments.

Findings

Outperforms existing inductive KG embedding models

Effectively embeds unseen entities and relations

Demonstrates superior performance on real-world datasets

Abstract

We study the knowledge extrapolation problem to embed new components (i.e., entities and relations) that come with emerging knowledge graphs (KGs) in the federated setting. In this problem, a model trained on an existing KG needs to embed an emerging KG with unseen entities and relations. To solve this problem, we introduce the meta-learning setting, where a set of tasks are sampled on the existing KG to mimic the link prediction task on the emerging KG. Based on sampled tasks, we meta-train a graph neural network framework that can construct features for unseen components based on structural information and output embeddings for them. Experimental results show that our proposed method can effectively embed unseen components and outperforms models that consider inductive settings for KGs and baselines that directly use conventional KG embedding methods.