KGATE : a Knowledge Graph Embedding Training Environment

2026-10-07

AI Summary

This study addresses the limitations of existing knowledge graph embedding libraries, including the lack of autoencoder support, insufficient maintenance, and incomparable results. To overcome these issues, this work proposes a modular Python framework built upon PyTorch Geometric and TorchKGE. The proposed approach enables flexible end-to-end assembly of encoders and decoders, integrates essential components such as initializers, loss functions, and negative sampling strategies, and incorporates built-in preprocessing mechanisms to prevent data leakage. Benchmark evaluations demonstrate that the framework achieves training efficiency comparable to state-of-the-art baselines while offering more comprehensive functionalities. Consequently, it significantly enhances cross-library experimental reproducibility and the reliability of model comparisons.

Abstract

Knowledge graph embedding (KGE) models encode the entities and relations of a knowledge graph into a low-dimensional latent space, enabling tasks such as classification or link prediction. Most KGE models follow an autoencoder architecture, in which an encoder projects the knowledge graph into the latent space and a decoder reconstruct it. Combining both encoder and decoder components is increasingly needed, yet existing libraries rarely support complete autoencoders, are often unmaintained, rely on undocumented default hyperparameters, and produce results that cannot be compared across libraries. Here we present KGATE (Knowledge Graph Autoencoder Training Environment), a modular Python library built on PyTorch Geometric and TorchKGE. KGATE lets users assemble initializers, encoders, decoders, losses, negative samplers, and evaluation metrics as building blocks, or plug in their own block. KGATE includes a preprocessing procedure that controls data leakage, a builtin training pipeline, and reproducibility by design. Benchmarks against six existing KGE libraries show that KGATE training time is comparable with the fastest libraries while offering a broader set of features.

Problem

• Knowledge Graph Embedding
• Autoencoder
• Reproducibility
• Benchmarking
• Library

Innovation

• Knowledge Graph Embedding
• Autoencoder
• Modular Library
• Reproducibility
• PyTorch Geometric