MPTSNet: Integrating Multiscale Periodic Local Patterns and Global Dependencies for Multivariate Time Series Classification

TL;DR

MPTSNet addresses multivariate time series classification by jointly modeling multiscale local patterns and global dependencies through FFT-based multiscale periodic decomposition and a PeriodicBlock consisting of an Inception-style Local Extractor and an Attention-based Global Capturer. It identifies main periods via FFT, reshapes data into multiple periodic scales, and adaptively aggregates scale-specific features using amplitude-guided weights to produce robust predictions with improved interpretability. Empirical results on the UEA benchmarks show state-of-the-art accuracy and favorable ranks against a wide range of general and MTSC-specific baselines, with ablations confirming the contribution of each component and the optimality of a moderate number of scales. The work advances MTSC by delivering a scalable, interpretable framework that leverages frequency-domain structure to enhance both performance and insight into temporal dynamics.

Abstract

Multivariate Time Series Classification (MTSC) is crucial in extensive practical applications, such as environmental monitoring, medical EEG analysis, and action recognition. Real-world time series datasets typically exhibit complex dynamics. To capture this complexity, RNN-based, CNN-based, Transformer-based, and hybrid models have been proposed. Unfortunately, current deep learning-based methods often neglect the simultaneous construction of local features and global dependencies at different time scales, lacking sufficient feature extraction capabilities to achieve satisfactory classification accuracy. To address these challenges, we propose a novel Multiscale Periodic Time Series Network (MPTSNet), which integrates multiscale local patterns and global correlations to fully exploit the inherent information in time series. Recognizing the multi-periodicity and complex variable correlations in time series, we use the Fourier transform to extract primary periods, enabling us to decompose data into multiscale periodic segments. Leveraging the inherent strengths of CNN and attention mechanism, we introduce the PeriodicBlock, which adaptively captures local patterns and global dependencies while offering enhanced interpretability through attention integration across different periodic scales. The experiments on UEA benchmark datasets demonstrate that the proposed MPTSNet outperforms 21 existing advanced baselines in the MTSC tasks.

Figures (4)

• Figure 1: Varying correlations between the subsequences of two variables across different periodic scales in MTS data.
• Figure 2: Multiscale periodic feature analysis of time series data using FFT. By decomposing time series into multiple frequency components, it reveals both local patterns and global dependencies across different periodic scales.
• Figure 3: Overview of the Multiscale Periodic Time Series Network (MPTSNet). The model processes multivariate time series data through FFT to identify the main periods. The data is then segmented into multiscale periodic components, where Local Extractor and Global Capturer in PeriodicBlock extract features sequentially at different scales. The features are aggregated by corresponding amplitudes and passed through the Classification head for final prediction.
• Figure 4: Interpretability visualization of the MPTSNet model on the Epilepsy dataset. Left: attention maps at different periodic scales (single example); Right: composite attention maps by weighted summing the individual scale maps. MPTSNet captures varying local patterns across periodic scales while achieving enhanced interpretability by the integration of multi-scale attention.