D-PAD: Deep-Shallow Multi-Frequency Patterns Disentangling for Time Series Forecasting
This work addresses a specific bottleneck in time series forecasting for applications requiring accurate pattern separation, representing an incremental advance over existing decomposition-based methods.
The paper tackles the problem of disentangling mixed-frequency patterns in time series forecasting by proposing D-PAD, a neural network that combines shallow decomposition with deep progressive extraction, achieving state-of-the-art performance with average improvements of 9.48% in MSE and 7.15% in MAE over baselines on seven datasets.
In time series forecasting, effectively disentangling intricate temporal patterns is crucial. While recent works endeavor to combine decomposition techniques with deep learning, multiple frequencies may still be mixed in the decomposed components, e.g., trend and seasonal. Furthermore, frequency domain analysis methods, e.g., Fourier and wavelet transforms, have limitations in resolution in the time domain and adaptability. In this paper, we propose D-PAD, a deep-shallow multi-frequency patterns disentangling neural network for time series forecasting. Specifically, a multi-component decomposing (MCD) block is introduced to decompose the series into components with different frequency ranges, corresponding to the "shallow" aspect. A decomposition-reconstruction-decomposition (D-R-D) module is proposed to progressively extract the information of frequencies mixed in the components, corresponding to the "deep" aspect. After that, an interaction and fusion (IF) module is used to further analyze the components. Extensive experiments on seven real-world datasets demonstrate that D-PAD achieves the state-of-the-art performance, outperforming the best baseline by an average of 9.48% and 7.15% in MSE and MAE, respectively.