Knowledge Graph (KG) completion is an important task that greatly benefits knowledge discovery in many fields (e.g. biomedical research). In recent years, learning KG embeddings to perform this task has received considerable attention. Despite the success of KG embedding methods, they predominantly use negative sampling, resulting in increased computational complexity as well as biased predictions due to the closed world assumption. To overcome these limitations, we propose \textbf{KG-NSF}, a negative sampling-free framework for learning KG embeddings based on the cross-correlation matrices of embedding vectors. It is shown that the proposed method achieves comparable link prediction performance to negative sampling-based methods while converging much faster.

翻译：知识图(KG)的完成是一项重要任务,它极大地有利于许多领域(例如生物医学研究)的知识发现。近年来,学习KG嵌入来完成这项任务的工作受到相当重视。尽管KG嵌入方法取得了成功,但它们主要使用负抽样方法,导致计算复杂性增加,而且由于封闭世界的假设而作出偏颇的预测。为了克服这些限制,我们提议了“textbf{KG-NSF}”这一不设抽样的负面框架,用于学习KG嵌入,其基础是嵌入矢量的交叉关系矩阵。这表明,拟议方法在将预测性能与负抽样方法相挂钩的同时,更快地融合。