[2504.07654] ms-Mamba: Multi-scale Mamba for Time-Series Forecasting

Computer Science > Machine Learning arXiv:2504.07654 (cs) [Submitted on 10 Apr 2025 (v1), last revised 5 Mar 2026 (this version, v2)] Title:ms-Mamba: Multi-scale Mamba for Time-Series Forecasting Authors:Yusuf Meric Karadag, Ismail Talaz, Ipek Gursel Dino, Sinan Kalkan View a PDF of the paper titled ms-Mamba: Multi-scale Mamba for Time-Series Forecasting, by Yusuf Meric Karadag and 3 other authors View PDF HTML (experimental) Abstract:The problem of Time-series Forecasting is generally addressed by recurrent, Transformer-based and the recently proposed Mamba-based architectures. However, existing architectures generally process their input at a single temporal scale, which may be sub-optimal for many tasks where information changes over multiple time scales. In this paper, we introduce a novel architecture called Multi-scale Mamba (ms-Mamba) to address this gap. ms-Mamba incorporates multiple temporal scales by using multiple Mamba blocks with different sampling rates ($\Delta$s). Our experiments on many benchmarks demonstrate that ms-Mamba outperforms state-of-the-art approaches, including the recently proposed Transformer-based and Mamba-based models. For example, on the Solar-Energy dataset, ms-Mamba outperforms its closest competitor S-Mamba (0.229 vs. 0.240 in terms of mean-squared error) while using fewer parameters (3.53M vs. 4.77M), less memory (13.46MB vs. 18.18MB), and less operations (14.93G vs. 20.53G MACs), averaged across four forecast lengths. Codes and mode...