FMMVCC: Fuzzy Mamba-based Multi-View Contrastive Clustering for Univariate Time Series

2026-07-08

AI Summary

This work addresses the challenges of modeling long-range dependencies and high computational costs in unsupervised time series clustering by introducing, for the first time, the Mamba state space model to this task. The proposed approach integrates fuzzy clustering with multi-view self-supervised learning, leveraging temporal masking and data augmentation strategies to effectively capture long-term dependencies while maintaining linear computational complexity. Extensive experiments across 15 benchmark datasets demonstrate that the method achieves state-of-the-art performance, yielding the best results on 29 out of 60 evaluation metrics and attaining the highest average rank across all scenarios, significantly outperforming existing methods.

Abstract

In many realistic scenarios, large volumes of time series data are generated with limited or expensive annotations. This limitation makes supervised learning methods difficult to apply and leads to the use of unsupervised approaches capable of discovering meaningful structures directly from raw data. Clustering therefore plays a crucial role in organizing time series into groups that share similar temporal patterns, enabling exploratory analysis and downstream tasks without requiring manual labeling. However, existing deep clustering methods often struggle to capture long-range temporal dependencies or rely on architectures with high computational cost. This paper introduces FMMVCC, a Mamba-based deep clustering framework for time series that leverages state space sequence modeling to efficiently learn temporal representations with linear complexity. Additionally, it utilizes multi-view self-supervised learning with temporal masking and augmentations. Experimental evaluation in 15 benchmark datasets proves that FMMVCC consistently outperforms state-of-the-art baselines, achieving the best overall performance in 29 of 60 total metric evaluations and the highest average rank in all tested scenarios.

Problem

• time series clustering
• long-range dependencies
• unsupervised learning
• computational efficiency
• deep clustering

Innovation

• Mamba
• state space model
• multi-view contrastive learning
• time series clustering
• self-supervised learning