Biography

I am pursuing my M.S. degree in College of Control Science and Engineering, Zhejiang University, Hangzhou, China. My major research lies at data mining and deep learning. I am currently working on general time series models.

Research and Interests

• Data Mining
• Deep Learning

Publications

• Changning Wu, Gao Wu, Rongyao Cai, Yong Liu, and Kexin Zhang. KFS: KAN Based Adaptive Frequency Selection Learning Architecture for Long-term Time Series Forecasting. Advanced Engineering Informatics, 72:104442, 2026.
  [BibTeX] [Abstract] [DOI] [PDF]

  Multi-scale decomposition architectures have emerged as predominant methodologies in time series forecasting. However, real-world time series exhibit noise interference across different scales, while heterogeneous information distribution among frequency components at varying scales leads to suboptimal multi-scale representation. Inspired by Kolmogorov-Arnold Networks (KAN) and Parseval’s theorem, we propose a KAN based adaptive frequency Selection learning architecture (KFS) to address these challenges. This framework tackles prediction challenges stemming from cross-scale noise interference and complex pattern modeling through its FreK module, which performs energy-distribution-based dominant frequency selection in the spectral domain. Simultaneously, KAN enables sophisticated pattern representation while timestamp embedding alignment synchronizes temporal representations across scales. The feature mixing module then fuses scale-specific patterns with aligned temporal features. Extensive experiments across multiple real-world time series datasets demonstrate that KFS achieves state-of-the-art performance as a simple yet effective architecture. Our code is available on this website: https://github.com/wcnExplosion/KFS-main.

  @article{wu2026kfs,
  title = {KFS: KAN Based Adaptive Frequency Selection Learning Architecture for Long-term Time Series Forecasting},
  author = {Changning Wu and Gao Wu and Rongyao Cai and Yong Liu and Kexin Zhang},
  year = 2026,
  journal = {Advanced Engineering Informatics},
  volume = 72,
  pages = {104442},
  doi = {10.1016/j.aei.2026.104442},
  abstract = {Multi-scale decomposition architectures have emerged as predominant methodologies in time series forecasting.
  However, real-world time series exhibit noise interference across different scales, while heterogeneous information distribution among frequency components at varying scales leads to suboptimal multi-scale representation.
  Inspired by Kolmogorov-Arnold Networks (KAN) and Parseval’s theorem, we propose a KAN based adaptive
  frequency Selection learning architecture (KFS) to address these challenges. This framework tackles prediction
  challenges stemming from cross-scale noise interference and complex pattern modeling through its FreK
  module, which performs energy-distribution-based dominant frequency selection in the spectral domain.
  Simultaneously, KAN enables sophisticated pattern representation while timestamp embedding alignment
  synchronizes temporal representations across scales. The feature mixing module then fuses scale-specific
  patterns with aligned temporal features. Extensive experiments across multiple real-world time series datasets
  demonstrate that KFS achieves state-of-the-art performance as a simple yet effective architecture. Our code is
  available on this website: https://github.com/wcnExplosion/KFS-main.}
  }