arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2101.05151cs.LGcs.AI
收藏

基于神经ODE的时态知识图预测

Temporal Knowledge Graph Forecasting with Neural ODE

Institute of Informatics, LMU Munich(慕尼黑大学信息学研究所);Corporate Technology, Siemens AG(西门子股份公司企业技术部);Technical University of Munich(慕尼黑工业大学)

Zhen Han, Zifeng Ding, Yunpu Ma, Yujia Gu, Volker Tresp

摘要:

在时间知识图谱(KG)上推断未来链接的兴趣越来越大。虽然时间KG上的链接随时间连续变化,但现有方法在离散状态空间中对时间KG进行建模。为此,我们通过将神经常微分方程(ODE)的思想扩展到多关系图卷积网络,提出了一种新型的连续统模型。该模型保持了动态多关系图数据的连续性,并将时间和结构信息编码为连续时间动态嵌入。此外,应用新颖的图过渡层来捕获动态图上的过渡,即,边缘形成和溶解。我们在五个基准数据集上进行了大量的时间KG推理实验,显示了我们的模型在未来链路预测任务上的优越性能。

英文摘要:

There has been an increasing interest in inferring future links on temporal knowledge graphs (KG). While links on temporal KGs vary continuously over time, the existing approaches model the temporal KGs in discrete state spaces. To this end, we propose a novel continuum model by extending the idea of neural ordinary differential equations (ODEs) to multi-relational graph convolutional networks. The proposed model preserves the continuous nature of dynamic multi-relational graph data and encodes both temporal and structural information into continuous-time dynamic embeddings. In addition, a novel graph transition layer is applied to capture the transitions on the dynamic graph, i.e., edge formation and dissolution. We perform extensive experiments on five benchmark datasets for temporal KG reasoning, showing our model's superior performance on the future link forecasting task.

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。

↑