FAIM: Frequency-Aware Interactive Mamba for Time Series Classification

2025-11-26

AI Summary

To address key challenges in time-series classification (TSC)—including high computational overhead, poor noise robustness, and overfitting under small-sample conditions—this paper proposes a lightweight frequency-aware interactive Mamba architecture. Methodologically: (i) an adaptive frequency-domain filtering block is designed, leveraging Fourier transform and learnable thresholds for robust noise suppression; (ii) an interactive Mamba block is introduced, integrating global-local semantic coupling to enhance multi-granularity temporal modeling; (iii) self-supervised pretraining is incorporated to improve generalization. Evaluated on multiple benchmark datasets, the method achieves significant improvements over state-of-the-art approaches, particularly excelling in high-noise and few-shot scenarios while maintaining high efficiency. Experimental results validate the effectiveness of synergistically combining frequency-domain priors with state-space modeling for robust and efficient TSC.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Time series classification (TSC) is crucial in numerous real-world applications, such as environmental monitoring, medical diagnosis, and posture recognition. TSC tasks require models to effectively capture discriminative information for accurate class identification. Although deep learning architectures excel at capturing temporal dependencies, they often suffer from high computational cost, sensitivity to noise perturbations, and susceptibility to overfitting on small-scale datasets. To address these challenges, we propose FAIM, a lightweight Frequency-Aware Interactive Mamba model. Specifically, we introduce an Adaptive Filtering Block (AFB) that leverages Fourier Transform to extract frequency-domain features from time series data. The AFB incorporates learnable adaptive thresholds to dynamically suppress noise and employs element-wise coupling of global and local semantic adaptive filtering, enabling in-depth modeling of the synergy among different frequency components. Furthermore, we design an Interactive Mamba Block (IMB) to facilitate efficient multi-granularity information interaction, balancing the extraction of fine-grained discriminative features and comprehensive global contextual information, thereby endowing FAIM with powerful and expressive representations for TSC tasks. Additionally, we incorporate a self-supervised pre-training mechanism to enhance FAIM's understanding of complex temporal patterns and improve its robustness across various domains and high-noise scenarios. Extensive experiments on multiple benchmarks demonstrate that FAIM consistently outperforms existing state-of-the-art (SOTA) methods, achieving a superior trade-off between accuracy and efficiency and exhibits outstanding performance.

Problem

• Addresses high computational cost in time series classification
• Mitigates sensitivity to noise and overfitting in small datasets
• Enhances feature extraction and robustness across diverse domains

Innovation

• Adaptive Filtering Block with Fourier Transform for noise suppression
• Interactive Mamba Block for multi-granularity information interaction
• Self-supervised pre-training to enhance robustness across domains