MS-DFTVNet:A Long-Term Time Series Prediction Method Based on Multi-Scale Deformable Convolution

TL;DR

This paper introduces MS-DFTVNet, a novel multi-scale deformable convolutional framework that effectively captures complex temporal dependencies for long-term time series prediction, outperforming existing Transformer and MLP models.

Contribution

The paper presents a new multi-scale reshape module and a context-aware deformable convolution mechanism, advancing convolutional approaches for long-term time series forecasting.

Findings

Achieves about 7.5% average improvement across six datasets

Outperforms strong baseline models significantly

Sets new state-of-the-art results in long-term forecasting

Abstract

Research on long-term time series prediction has primarily relied on Transformer and MLP models, while the potential of convolutional networks in this domain remains underexplored. To address this, we propose a novel multi-scale time series reshape module that effectively captures cross-period patch interactions and variable dependencies. Building on this, we develop MS-DFTVNet, the multi-scale 3D deformable convolutional framework tailored for long-term forecasting. Moreover, to handle the inherently uneven distribution of temporal features, we introduce a context-aware dynamic deformable convolution mechanism, which further enhances the model's ability to capture complex temporal patterns. Extensive experiments demonstrate that MS-DFTVNet not only significantly outperforms strong baselines but also achieves an average improvement of about 7.5% across six public datasets, setting new state-of-the-art results.