用图神经流建模不规则采样时间序列的系统级交互关系
Graph Neural Flows for Unveiling Systemic Interactions Among Irregularly Sampled Time Series
- 构建有向无环图表示组件间条件依赖关系
- 在连续时间模型中联合学习图结构与ODE解曲线
- 适合处理不规则采样、需揭示因果交互的时序数据
自然界中普遍存在相互作用的系统。若独立分析其组成成分，难以准确预测系统动态。本文提出一种基于图的模型，通过有向无环图建模系统成分间的条件依赖（一种因果表示），并将其与参数化常微分方程（ODE）解曲线的连续时间模型联合学习。该方法称为图神经流，在时间序列分类和预测等任务上显著优于非图基方法及未建模条件依赖的图基方法，验证了其有效性。
原文摘要 · Abstract (English)
Interacting systems are prevalent in nature. It is challenging to accurately predict the dynamics of the system if its constituent components are analyzed independently. We develop a graph-based model that unveils the systemic interactions of time series observed at irregular time points, by using a directed acyclic graph to model the conditional dependencies (a form of causal notation) of the system components and learning this graph in tandem with a continuous-time model that parameterizes the solution curves of ordinary differential equations (ODEs). Our technique, a graph neural flow, leads to substantial enhancements over non-graph-based methods, as well as graph-based methods without the modeling of conditional dependencies. We validate our approach on several tasks, including time series classification and forecasting, to demonstrate its efficacy.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。