Omni-Dimensional Frequency Learner for General Time Series Analysis
This work addresses a broad problem in time series analysis for researchers and practitioners by providing a novel frequency-based approach that outperforms existing methods, though it appears incremental as it builds on prior frequency domain techniques.
The paper tackled the problem of frequency-based methods underperforming compared to time domain methods in general time series analysis by proposing the Omni-Dimensional Frequency Learner (ODFL), which achieved state-of-the-art results across five tasks including forecasting, imputation, classification, and anomaly detection.
Frequency domain representation of time series feature offers a concise representation for handling real-world time series data with inherent complexity and dynamic nature. However, current frequency-based methods with complex operations still fall short of state-of-the-art time domain methods for general time series analysis. In this work, we present Omni-Dimensional Frequency Learner (ODFL) model based on a in depth analysis among all the three aspects of the spectrum feature: channel redundancy property among the frequency dimension, the sparse and un-salient frequency energy distribution among the frequency dimension, and the semantic diversity among the variable dimension. Technically, our method is composed of a semantic-adaptive global filter with attention to the un-salient frequency bands and partial operation among the channel dimension. Empirical results show that ODFL achieves consistent state-of-the-art in five mainstream time series analysis tasks, including short- and long-term forecasting, imputation, classification, and anomaly detection, offering a promising foundation for time series analysis.