FedMKGC: Privacy-Preserving Federated Multilingual Knowledge Graph Completion

TL;DR

This paper introduces FedMKGC, a federated learning framework for multilingual knowledge graph completion that preserves privacy by avoiding raw data sharing and entity alignment, while effectively leveraging multilingual knowledge.

Contribution

It proposes a novel federated learning approach that implicitly aggregates multilingual knowledge graphs using language models, eliminating the need for raw data exchange or entity alignment.

Findings

Achieves comparable performance to state-of-the-art models

Significantly improves multilingual knowledge graph completion

Maintains privacy by not sharing raw data or alignments

Abstract

Knowledge graph completion (KGC) aims to predict missing facts in knowledge graphs (KGs), which is crucial as modern KGs remain largely incomplete. While training KGC models on multiple aligned KGs can improve performance, previous methods that rely on transferring raw data among KGs raise privacy concerns. To address this challenge, we propose a new federated learning framework that implicitly aggregates knowledge from multiple KGs without demanding raw data exchange and entity alignment. We treat each KG as a client that trains a local language model through textbased knowledge representation learning. A central server then aggregates the model weights from clients. As natural language provides a universal representation, the same knowledge thus has similar semantic representations across KGs. As such, the aggregated language model can leverage complementary knowledge from multilingual KGs without demanding raw user data sharing. Extensive experiments on a benchmark dataset demonstrate that our method substantially improves KGC on multilingual KGs, achieving comparable performance to state-of-the-art alignment-based models without requiring any labeled alignments or raw user data sharing. Our codes will be publicly available.