KFS: KAN based adaptive Frequency Selection learning architecture for long term time series forecasting
This addresses forecasting accuracy for real-world time series applications, representing an incremental improvement over existing multi-scale decomposition methods.
The paper tackles the problem of noise interference and heterogeneous information distribution in multi-scale time series forecasting by proposing KFS, a KAN-based adaptive frequency selection architecture. Experiments show it achieves state-of-the-art performance on multiple real-world datasets.
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 KT achieves state-of-the-art performance as a simple yet effective architecture.