MSTN: A Lightweight and Fast Model for General TimeSeries Analysis

TL;DR

MSTN is a lightweight, multi-scale neural network architecture designed for versatile and efficient time series analysis, outperforming existing methods across various benchmarks.

Contribution

The paper introduces MSTN, a hybrid neural model that captures multi-scale temporal features dynamically, improving adaptability and performance in time series tasks.

Findings

Achieves state-of-the-art results on 21 of 27 datasets.

Remains lightweight with around 0.40M to 1.06M parameters.

Enables low-latency inference suitable for resource-constrained environments.

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 a single dense layer to dynamically reweight and fuse multi-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 imputation, long-term forecasting, classification, and cross-dataset generalization, MSTN achieves state-of-the-art performance, establishing new best results on 21 of 27 datasets, while remaining lightweight (~0.40M params for MSTN-BiLSTM and ~1.06M for MSTN-Transformer) and suitable for low-latency inference (<1 sec, often in milliseconds), resource-constrained deployment.