arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.02707cs.LGcs.AI
收藏

SnapE——训练链接预测模型的快照集成

SnapE -- Training Snapshot Ensembles of Link Prediction Models

University of Mannheim(曼海姆大学);Amazon Web Services(亚马逊网络服务)

Ali Shaban, Heiko Paulheim

AI总结:

针对知识图谱链接预测缺乏显式负样本的问题,提出SnapE快照集成方法,利用先前快照模型迭代生成负样本,在不增加训练时间的前提下稳定提升性能。

摘要:

快照集成已广泛应用于各个预测领域。它们允许以训练单个预测模型的成本训练一个预测模型集成。众所周知,它们通过创建一组多样化的基础模型来产生更稳健的预测。在本文中,我们提出了一种将快照集成思想迁移到知识图谱中链接预测模型的方法。此外,由于知识图谱中的链接预测是一种没有显式负样本的设置,我们提出了一种新颖的训练循环,该循环使用先前的快照模型迭代地创建负样本。在四个数据集上使用四个基础模型进行的评估表明,该方法在保持训练时间不变的同时,始终优于单一模型方法。

英文摘要:

Snapshot ensembles have been widely used in various fields of prediction. They allow for training an ensemble of prediction models at the cost of training a single one. They are known to yield more robust predictions by creating a set of diverse base models. In this paper, we introduce an approach to transfer the idea of snapshot ensembles to link prediction models in knowledge graphs. Moreover, since link prediction in knowledge graphs is a setup without explicit negative examples, we propose a novel training loop that iteratively creates negative examples using previous snapshot models. An evaluation with four base models across four datasets shows that this approach constantly outperforms the single model approach, while keeping the training time constant.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑