MSTN: A Lightweight and Fast Model for General Time-Series Analysis

TMLR Paper7524 Authors

Abstract: Real-world time series often exhibit strong non-stationarity, complex nonlinear dynamics, and behavior expressed across multiple temporal scales, from rapid local fluctuations to slow-evolving long-range trends. However, many contemporary architectures impose rigid, fixed-scale structural priors---such as patch-based tokenization, predefined receptive fields, or frozen backbone encoders---which can over-regularize temporal dynamics and limit adaptability to abrupt high-magnitude events. To handle this, we introduce the Multi-scale Temporal Network (MSTN), a hybrid neural architecture grounded in an Early Temporal Aggregation principle. MSTN integrates three complementary components: (i) a multi-scale convolutional encoder that captures fine-grained local structure; (ii) a sequence modeling module that learns long-range dependencies through either recurrent or attention-based mechanisms; and (iii) a self-gated fusion stage incorporating squeeze--excitation and multi-head attention to dynamically modulate cross-scale representations. This design enables MSTN to flexibly model temporal patterns spanning milliseconds to extended horizons, while avoiding the computational burden typically associated with long-context models. Across extensive benchmarks covering forecasting, imputation, classification, and cross-dataset generalization, MSTN achieves state-of-the-art performance, establishing new best results on 24 of 32 datasets, while remaining lightweight (≈ 1M params) and suitable for low-latency (<1 sec, often in milliseconds), resource-constrained deployment.

Submission Type: Long submission (more than 12 pages of main content)

Assigned Action Editor: ~Markus_Lange-Hegermann1

Submission Number: 7524