A Unified Frequency Domain Decomposition Framework for Interpretable and Robust Time Series Forecasting
This work addresses the challenge of interpretable and robust forecasting for time series data, representing an incremental improvement over existing deep learning methods.
The paper tackled the problem of limited interpretability and robustness in time series forecasting by proposing FIRE, a unified frequency domain decomposition framework, which outperformed state-of-the-art models on long-term forecasting benchmarks with superior predictive performance and enhanced interpretability.
Current approaches for time series forecasting, whether in the time or frequency domain, predominantly use deep learning models based on linear layers or transformers. They often encode time series data in a black-box manner and rely on trial-and-error optimization solely based on forecasting performance, leading to limited interpretability and theoretical understanding. Furthermore, the dynamics in data distribution over time and frequency domains pose a critical challenge to accurate forecasting. We propose FIRE, a unified frequency domain decomposition framework that provides a mathematical abstraction for diverse types of time series, so as to achieve interpretable and robust time series forecasting. FIRE introduces several key innovations: (i) independent modeling of amplitude and phase components, (ii) adaptive learning of weights of frequency basis components, (iii) a targeted loss function, and (iv) a novel training paradigm for sparse data. Extensive experiments demonstrate that FIRE consistently outperforms state-of-the-art models on long-term forecasting benchmarks, achieving superior predictive performance and significantly enhancing interpretability of time series