Learning to Forecast on Multiple Spatiotemporal Hierarchies

Key points

• In the associated deep learning architectures, the hierarchy determines which aggregate series are available and how they are processed together with individual series (e.g., through reconciliation or message-passing operators).1
• In this work, we propose a graph deep learning framework that leverages variations across hierarchies as multi-task regularization.1
• We show that jointly learning across these structures can reduce over-specialization to any single hierarchy and improve forecasting performance.1
• Making forecasts coherent with the aggregation constraints for each sampled hierarchy is computationally expensive: to complement the framework, we propose a novel operator-based on a top-down approach-that produces coherent forecasts with linear time and space complexity w.r.t. the hierarchy size.1

Sources (1)

• [1]Learning to Forecast on Multiple Spatiotemporal Hierarchies
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 15, 07:29 AM

  Hierarchical forecasting exploits aggregation structures that expose multiple views of the same underlying spatiotemporal forecasting problem, while ensuring coherence among forecasts.

  In the associated deep learning architectures, the hierarchy determines which aggregate series are available and how they are processed together with individual series (e.g., through reconciliation or message-passing operators).

Extractive summary: sentences quoted from the sources.