FreqMoE: Enhancing Time Series Forecasting through Frequency Decomposition Mixture of Experts

TL;DR

FreqMoE introduces a novel frequency decomposition mixture-of-experts model that dynamically processes different frequency bands of time series data, leading to improved forecasting accuracy and efficiency over existing methods.

Contribution

The paper proposes a dynamic frequency decomposition approach with specialized experts and a gating mechanism, enhancing long-term time series forecasting accuracy and efficiency.

Findings

Outperforms state-of-the-art models on 51 out of 70 metrics

Reduces model parameters to under 50,000, improving efficiency

Achieves superior forecasting accuracy across diverse datasets

Abstract

Long-term time series forecasting is essential in areas like finance and weather prediction. Besides traditional methods that operate in the time domain, many recent models transform time series data into the frequency domain to better capture complex patterns. However, these methods often use filtering techniques to remove certain frequency signals as noise, which may unintentionally discard important information and reduce prediction accuracy. To address this, we propose the Frequency Decomposition Mixture-of-Experts (FreqMoE) model, which dynamically decomposes time series data into frequency bands, each processed by a specialized expert. A gating mechanism adjusts the importance of each output of expert based on frequency characteristics, and the aggregated results are fed into a prediction module that iteratively refines the forecast using residual connections. Our experiments demonstrate that FreqMoE outperforms state-of-the-art models, achieving the best performance on 51 out of 70 metrics across all tested datasets, while significantly reducing the number of required parameters to under 50k, providing notable efficiency advantages. Code is available at: https://github.com/sunbus100/FreqMoE-main