TimeCF: A TimeMixer-Based Model with adaptive Convolution and Sharpness-Aware Minimization Frequency Domain Loss for long-term time seris forecasting

TL;DR

TimeCF addresses long-horizon time series forecasting challenges by coupling a TimeMixer-based decomposition–mixing backbone with adaptive multi-scale convolution and a frequency-domain loss. The PDMC module enables scale-aware, receptive-field–dependent information aggregation, while SAMFre decouples label autocorrelation and regularizes the loss via a Fourier-domain component. Empirical results on Weather, ETTh1/ETTh2, ETTm1/ETTm2, and Electricity demonstrate strong performance and parameter efficiency, with ablations confirming the necessity of both adaptive convolution and SAMFre. This approach offers a practical, scalable solution for real-world, long-term forecasting where autocorrelation and non-stationarity are prominent concerns.

Abstract

Recent studies have shown that by introducing prior knowledge, multi-scale analysis of complex and non-stationary time series in real environments can achieve good results in the field of long-term forecasting. However, affected by channel-independent methods, models based on multi-scale analysis may produce suboptimal prediction results due to the autocorrelation between time series labels, which in turn affects the generalization ability of the model. To address this challenge, we are inspired by the idea of sharpness-aware minimization and the recently proposed FreDF method and design a deep learning model TimeCF for long-term time series forecasting based on the TimeMixer, combined with our designed adaptive convolution information aggregation module and Sharpness-Aware Minimization Frequency Domain Loss (SAMFre). Specifically, TimeCF first decomposes the original time series into sequences of different scales. Next, the same-sized convolution modules are used to adaptively aggregate information of different scales on sequences of different scales. Then, decomposing each sequence into season and trend parts and the two parts are mixed at different scales through bottom-up and top-down methods respectively. Finally, different scales are aggregated through a Feed-Forward Network. What's more, extensive experimental results on different real-world datasets show that our proposed TimeCF has excellent performance in the field of long-term forecasting.

Figures (1)

• Figure 1: TimeCF Architecture