Research

Paper

AI LLM March 10, 2026

FreqCycle: A Multi-Scale Time-Frequency Analysis Method for Time Series Forecasting

Authors

Boya Zhang, Shuaijie Yin, Huiwen Zhu, Xing He

Abstract

Mining time-frequency features is critical for time series forecasting. Existing research has predominantly focused on modeling low-frequency patterns, where most time series energy is concentrated. The overlooking of mid to high frequency continues to limit further performance gains in deep learning models. We propose FreqCycle, a novel framework integrating: (i) a Filter-Enhanced Cycle Forecasting (FECF) module to extract low-frequency features by explicitly learning shared periodic patterns in the time domain, and (ii) a Segmented Frequency-domain Pattern Learning (SFPL) module to enhance mid to high frequency energy proportion via learnable filters and adaptive weighting. Furthermore, time series data often exhibit coupled multi-periodicity, such as intertwined weekly and daily cycles. To address coupled multi-periodicity as well as long lookback window challenges, we extend FreqCycle hierarchically into MFreqCycle, which decouples nested periodic features through cross-scale interactions. Extensive experiments on seven diverse domain benchmarks demonstrate that FreqCycle achieves state-of-the-art accuracy while maintaining faster inference speeds, striking an optimal balance between performance and efficiency.

Metadata

arXiv ID: 2603.09661
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-03-10
Fetched: 2026-03-11 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.09661v1</id>\n    <title>FreqCycle: A Multi-Scale Time-Frequency Analysis Method for Time Series Forecasting</title>\n    <updated>2026-03-10T13:34:36Z</updated>\n    <link href='https://arxiv.org/abs/2603.09661v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.09661v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Mining time-frequency features is critical for time series forecasting. Existing research has predominantly focused on modeling low-frequency patterns, where most time series energy is concentrated. The overlooking of mid to high frequency continues to limit further performance gains in deep learning models. We propose FreqCycle, a novel framework integrating: (i) a Filter-Enhanced Cycle Forecasting (FECF) module to extract low-frequency features by explicitly learning shared periodic patterns in the time domain, and (ii) a Segmented Frequency-domain Pattern Learning (SFPL) module to enhance mid to high frequency energy proportion via learnable filters and adaptive weighting. Furthermore, time series data often exhibit coupled multi-periodicity, such as intertwined weekly and daily cycles. To address coupled multi-periodicity as well as long lookback window challenges, we extend FreqCycle hierarchically into MFreqCycle, which decouples nested periodic features through cross-scale interactions. Extensive experiments on seven diverse domain benchmarks demonstrate that FreqCycle achieves state-of-the-art accuracy while maintaining faster inference speeds, striking an optimal balance between performance and efficiency.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <published>2026-03-10T13:34:36Z</published>\n    <arxiv:comment>18 pages, 17 figures, accepted to AAAI 2026. Code available at https://github.com/boya-zhang-ai/FreqCycle</arxiv:comment>\n    <arxiv:primary_category term='cs.LG'/>\n    <author>\n      <name>Boya Zhang</name>\n    </author>\n    <author>\n      <name>Shuaijie Yin</name>\n    </author>\n    <author>\n      <name>Huiwen Zhu</name>\n    </author>\n    <author>\n      <name>Xing He</name>\n    </author>\n  </entry>"
}