FITS: Modeling Time Series with $10k$ Parameters
This enables lightweight training and deployment on edge devices for applications like forecasting and anomaly detection, though it appears incremental as it builds on frequency-domain methods.
The paper tackles time series analysis by introducing FITS, a model that operates in the complex frequency domain and discards high-frequency components, achieving performance comparable to state-of-the-art models with only about 10k parameters.
In this paper, we introduce FITS, a lightweight yet powerful model for time series analysis. Unlike existing models that directly process raw time-domain data, FITS operates on the principle that time series can be manipulated through interpolation in the complex frequency domain. By discarding high-frequency components with negligible impact on time series data, FITS achieves performance comparable to state-of-the-art models for time series forecasting and anomaly detection tasks, while having a remarkably compact size of only approximately $10k$ parameters. Such a lightweight model can be easily trained and deployed in edge devices, creating opportunities for various applications. The code is available in: \url{https://github.com/VEWOXIC/FITS}