Context-aware forecasting improves time-series predictions with manager-worker design
CATF: A Manager-Worker Framework for Context-Aware Time-Series Forecasting
Machine Learning
Summary
Forecasting future data points, like bike usage or weather conditions, can get tricky when many factors influence the outcome simultaneously. The authors created a new method named CATF that uses a 'manager' to decide which specialized 'worker' model should handle each prediction based on context like weather or calendar info. This way, each worker focuses on a specific pattern without needing manual splits, making predictions more accurate and easier to understand. Tests on various datasets showed CATF often beats other forecasting models, especially for complex data with changing conditions.
What this means in practice
- •For demand planners: Improve demand forecasts by routing inputs with calendar and contextual factors to specialized prediction models.
- •For energy grid operators: Enhance load forecasting accuracy by dynamically selecting specialized models based on weather and temporal context.
Authors
Yueqiao Chen, Dongyu Liu
Abstract
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.