Robust Knowledge Graph Embedding via Denoising

TL;DR

This paper introduces a denoising framework for knowledge graph embeddings that enhances robustness against noisy data and perturbations, utilizing energy-based modeling and certified robustness metrics.

Contribution

It presents a novel denoising approach for KGE models, connecting score matching with robustness training, and introduces certified robustness evaluation metrics.

Findings

Outperforms existing KGE methods on noisy datasets

Provides certified robustness metrics for KGE models

Demonstrates improved stability under embedding perturbations

Abstract

We focus on obtaining robust knowledge graph embedding under perturbation in the embedding space. To address these challenges, we introduce a novel framework, Robust Knowledge Graph Embedding via Denoising, which enhances the robustness of KGE models on noisy triples. By treating KGE methods as energy-based models, we leverage the established connection between denoising and score matching, enabling the training of a robust denoising KGE model. Furthermore, we propose certified robustness evaluation metrics for KGE methods based on the concept of randomized smoothing. Through comprehensive experiments on benchmark datasets, our framework consistently shows superior performance compared to existing state-of-the-art KGE methods when faced with perturbed entity embedding.