MPPN: Multi-Resolution Periodic Pattern Network For Long-Term Time Series Forecasting

TL;DR

This paper introduces MPPN, a novel deep learning architecture that captures intrinsic multi-resolution periodic patterns in long-term time series, improving forecasting accuracy by addressing noise and individual variate characteristics.

Contribution

The paper proposes MPPN, a multi-resolution network with a channel adaptive module and an entropy-based predictability measure, advancing long-term time series forecasting methods.

Findings

MPPN outperforms state-of-the-art methods on nine real-world benchmarks.

The entropy-based measure effectively evaluates the upper bound of prediction accuracy.

Multi-resolution pattern mining enhances the capture of intrinsic time series patterns.

Abstract

Long-term time series forecasting plays an important role in various real-world scenarios. Recent deep learning methods for long-term series forecasting tend to capture the intricate patterns of time series by decomposition-based or sampling-based methods. However, most of the extracted patterns may include unpredictable noise and lack good interpretability. Moreover, the multivariate series forecasting methods usually ignore the individual characteristics of each variate, which may affecting the prediction accuracy. To capture the intrinsic patterns of time series, we propose a novel deep learning network architecture, named Multi-resolution Periodic Pattern Network (MPPN), for long-term series forecasting. We first construct context-aware multi-resolution semantic units of time series and employ multi-periodic pattern mining to capture the key patterns of time series. Then, we propose a channel adaptive module to capture the perceptions of multivariate towards different patterns. In addition, we present an entropy-based method for evaluating the predictability of time series and providing an upper bound on the prediction accuracy before forecasting. Our experimental evaluation on nine real-world benchmarks demonstrated that MPPN significantly outperforms the state-of-the-art Transformer-based, decomposition-based and sampling-based methods for long-term series forecasting.