TSCMamba: Mamba Meets Multi-View Learning for Time Series Classification

TL;DR

TSCMamba introduces a multi-view, shift-equivariance aware approach for multivariate time series classification, integrating spectral, temporal, and global features with a novel sequence model and scanning scheme, leading to significant accuracy improvements.

Contribution

The paper proposes a novel multi-view method combining spectral and temporal features with the Mamba model and tango scanning to enhance TSC accuracy and robustness.

Findings

Achieved 4.01-6.45% accuracy improvements over TimesNet and TSLANet.

Effectively captures shift invariance and long-range dependencies.

Demonstrated robustness and scalability on benchmark datasets.

Abstract

Multivariate time series classification (TSC) is critical for various applications in fields such as healthcare and finance. While various approaches for TSC have been explored, important properties of time series, such as shift equivariance and inversion invariance, are largely underexplored by existing works. To fill this gap, we propose a novel multi-view approach to capture patterns with properties like shift equivariance. Our method integrates diverse features, including spectral, temporal, local, and global features, to obtain rich, complementary contexts for TSC. We use continuous wavelet transform to capture time-frequency features that remain consistent even when the input is shifted in time. These features are fused with temporal convolutional or multilayer perceptron features to provide complex local and global contextual information. We utilize the Mamba state space model for efficient and scalable sequence modeling and capturing long-range dependencies in time series. Moreover, we introduce a new scanning scheme for Mamba, called tango scanning, to effectively model sequence relationships and leverage inversion invariance, thereby enhancing our model's generalization and robustness. Experiments on two sets of benchmark datasets (10+20 datasets) demonstrate our approach's effectiveness, achieving average accuracy improvements of 4.01-6.45\% and 7.93\% respectively, over leading TSC models such as TimesNet and TSLANet.