[2603.19315] MSNet and LS-Net: Scalable Multi-Scale Multi-Representation Networks for Time Series Classification

Computer Science > Machine Learning arXiv:2603.19315 (cs) [Submitted on 14 Mar 2026] Title:MSNet and LS-Net: Scalable Multi-Scale Multi-Representation Networks for Time Series Classification Authors:Celal Alagöz, Mehmet Kurnaz, Farhan Aadil View a PDF of the paper titled MSNet and LS-Net: Scalable Multi-Scale Multi-Representation Networks for Time Series Classification, by Celal Alag\"oz and 2 other authors View PDF Abstract:Time series classification (TSC) performance depends not only on architectural design but also on the diversity of input representations. In this work, we propose a scalable multi-scale convolutional framework that systematically integrates structured multi-representation inputs for univariate time series. We introduce two architectures: MSNet, a hierarchical multi-scale convolutional network optimized for robustness and calibration, and LS-Net, a lightweight variant designed for efficiency-aware deployment. In addition, we adapt LiteMV -- originally developed for multivariate inputs -- to operate on multi-representation univariate signals, enabling cross-representation interaction. We evaluate all models across 142 benchmark datasets under a unified experimental protocol. Critical Difference analysis confirms statistically significant performance differences among the top models. Results show that LiteMV achieves the highest mean accuracy, MSNet provides superior probabilistic calibration (lowest NLL), and LS-Net offers the best efficiency-accuracy tr...