Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Rethinking the Power of Timestamps for Robust Time Series Forecasting: A Global-Local Fusion Perspective

Chengsen Wang, Qi Qi, Jingyu Wang, Haifeng Sun, Zirui Zhuang, Jinming Wu, Jianxin Liao

TL;DR

GLAFF adaptively adjusts the combined weights for global and local information, enabling seamless collaboration with any time series forecasting backbone, and significantly enhances the average performance of widely used mainstream forecasting models.

Abstract

Time series forecasting has played a pivotal role across various industries, including finance, transportation, energy, healthcare, and climate. Due to the abundant seasonal information they contain, timestamps possess the potential to offer robust global guidance for forecasting techniques. However, existing works primarily focus on local observations, with timestamps being treated merely as an optional supplement that remains underutilized. When data gathered from the real world is polluted, the absence of global information will damage the robust prediction capability of these algorithms. To address these problems, we propose a novel framework named GLAFF. Within this framework, the timestamps are modeled individually to capture the global dependencies. Working as a plugin, GLAFF adaptively adjusts the combined weights for global and local information, enabling seamless collaboration with any time series forecasting backbone. Extensive experiments conducted on nine real-world datasets demonstrate that GLAFF significantly enhances the average performance of widely used mainstream forecasting models by 12.5%, surpassing the previous state-of-the-art method by 5.5%.

Rethinking the Power of Timestamps for Robust Time Series Forecasting: A Global-Local Fusion Perspective

TL;DR

GLAFF adaptively adjusts the combined weights for global and local information, enabling seamless collaboration with any time series forecasting backbone, and significantly enhances the average performance of widely used mainstream forecasting models.

Abstract

Time series forecasting has played a pivotal role across various industries, including finance, transportation, energy, healthcare, and climate. Due to the abundant seasonal information they contain, timestamps possess the potential to offer robust global guidance for forecasting techniques. However, existing works primarily focus on local observations, with timestamps being treated merely as an optional supplement that remains underutilized. When data gathered from the real world is polluted, the absence of global information will damage the robust prediction capability of these algorithms. To address these problems, we propose a novel framework named GLAFF. Within this framework, the timestamps are modeled individually to capture the global dependencies. Working as a plugin, GLAFF adaptively adjusts the combined weights for global and local information, enabling seamless collaboration with any time series forecasting backbone. Extensive experiments conducted on nine real-world datasets demonstrate that GLAFF significantly enhances the average performance of widely used mainstream forecasting models by 12.5%, surpassing the previous state-of-the-art method by 5.5%.
Paper Structure (27 sections, 4 equations, 5 figures, 7 tables, 1 algorithm)

This paper contains 27 sections, 4 equations, 5 figures, 7 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methodology
  4. Overview
  5. Attention-based Mapper
  6. Robust Denormalizer
  7. Adaptive Combiner
  8. Experiment
  9. Experimental Setup
  10. Dataset
  11. Backbone
  12. Main Result
  13. Prediction Showcase
  14. Ablation Study
  15. Conclusion
  16. ...and 12 more sections

Figures (5)

  • Figure 1: The experimental results on Traffic dataset. (a) illustrates the outcomes of the ablation study on mainstream forecasting models and their variants. (b) depicts the visualization of traffic volume (upper), successful prediction case (lower right), and failed prediction case (lower left), respectively.
  • Figure 2: The overall architecture of GLAFF mainly consists of three primary components: Attention-based Mapper, Robust Denormalizer, and Adaptive Combiner.
  • Figure 3: The illustration of prediction showcases among GLAFF and mainstream baselines.
  • Figure 4: The forecasting errors for multivariate time series of hyperparameter analysis among different configurations for GLAFF. A lower outcome indicates a better prediction.
  • Figure 5: The complete illustration of prediction showcases among GLAFF and mainstream baselines.