[论文] CATF: A Manager-Worker Framework for Context-Aware Time-Series Forecas...

论文概要

研究领域: ML
作者: Yueqiao Chen, Dongyu Liu
发布时间: 2026-11-03
arXiv: 2611.05325

中文摘要

我们研究情境感知时间序列预测：单一目标变量由其自身历史与一组情境变量（如天气、日历信息）共同预测。随着情境愈发多样，单一模型难以捕捉所有模态，而集成学习与混合专家方法依赖手工数据划分或软路由，推高推理成本并损害可解释性。我们提出 CATF——一种基于 Manager-Worker 架构的情境感知时间序列预测框架：基于 Transformer 的 Manager 根据情境把每个输入硬路由到单一 Worker，训练与推理阶段都采用硬路由。Worker 因此无需手工划分即可专精化，任意单模型预测器都可充当 Worker。在五个富情境数据集、四个预测 horizon、两种归一化方案上，CATF 在 200 组对比中有 162 组优于其独立对应模型；最佳 CATF 增强模型在 20 个设置中有 19 个超过最佳独立基线，在 20 个设置中有 15 个超过零样本 Time-MoE 基础模型。单车共享数据上的案例研究显示，Manager 按日历路由输入，每个 Worker 专精于不同的周模式。

原文摘要

We study context-aware time-series forecasting, in which a single target variable is predicted from its own history and a set of contextual variables, such as weather and calendar information. As the context becomes more diverse, a single model struggles to capture all of its regimes, while ensemble learning and Mixture-of-Experts approaches rely on manual data partitioning or soft routing, which increases the inference cost and hinders interpretability. We propose CATF, a Context-Aware Time-Series Forecasting framework built on a Manager-Worker architecture: a Transformer-based Manager routes each input to a single Worker based on the context, using hard routing during both training and inference. The Workers thus specialize without manual data partitioning, and any single-model forecaster can serve as a Worker. On five context-rich datasets, four horizons, and two normalization schemes, CATF improves over its standalone counterpart in 162 of 200 comparisons, the best CATF-enhanced model outperforms the best standalone baseline in 19 of 20 settings, and it outperforms the zero-shot Time-MoE foundation model in 15 of 20 settings. A case study on bike-sharing data shows that the Manager routes inputs by the calendar and that each Worker specializes in a different weekly pattern.

自动采集于 2026-11-05

#论文 #arXiv #AI #小凯