Multivariate Time Series Classification

Ask about this topic

Overview

The field studies how to assign labels to multichannel sequences, including activity recordings, medical signals, and industrial sensor data. Approaches range from statistical and frequency-domain features, shapelet and random-kernel methods such as ROCKET, to convolutional networks, transformers, state-space models, and graph-based methods that capture channel interactions. Researchers compare methods on archives such as UEA and UCR, while also addressing irregular sampling, missing values, few-shot learning, and interpretability. A central tension is whether complex models improve on efficient feature-based methods, and whether benchmark gains hold under shifts between subjects, sites, or deployment conditions and under statistically sound evaluation.

Latest papers

CardsList

38 papers4 papers in the last four weeks, level with the four weeks before. 0.0% of all new papers.

  1. MIHT: A Hoeffding Tree for Time Series Classification using Multiple Instance Learning

    Mar 23, 2026Aurora Esteban, Amelia Zafra, Sebastián VenturaInterpretable Time-Series ClassificationTime Series Classification

    Dept. of Computer Science and Artificial Intelligence, Andalusian Research Institute in Data Science and Computational Intelligence, University of Cordoba, Spain

  2. A Direct Classification Approach for Reliable Wind Ramp Event Forecasting under Severe Class Imbalance

    Mar 20, 2026Alejandro Morales-Hernández, Fabrizio De Caroa, Gian Marco Paldino +3Wind Power ForecastingClass-Imbalanced Learning

    Machine Learning Group, Universite Libre de Bruxelles, Brussels, 1050, Belgium · University of Sannio, Benevento, 82100, Italy · WEL Research Institute, Wavre, 1300, Belgium

  3. The Multiverse of Time Series Machine Learning: an Archive for Multivariate Time Series Classification

    Mar 20, 2026Matthew Middlehurst, Aiden Rushbrooke, Ali Ismail-Fawaz +6Time Series ClassificationMultivariate Time Series Classification

    University of Bradford · University of East Anglia · Université de Haute-Alsace +2

  4. Data-Local Autonomous LLM-Guided Neural Architecture Search for Multiclass Multimodal Time-Series Classification

    Mar 16, 2026Emil Hardarson, Luka Biedebach, Ómar Bessi Ómarsson +3Large Language Model-Guided NASMultivariate Time Series Classification

    Reykjavik University, Reykjavik, Iceland · University of Southampton, Southampton, United Kingdom

  5. Permutation-Equivariant 2D State Space Models: Theory and Canonical Architecture for Multivariate Time Series

    Mar 7, 2026Seungwoo Jeong, Heung-Il SukState Space ModelsMultivariate Time Series Forecasting

    Department of Artificial Intelligence, Korea University, Seoul 02841, Republic of Korea

  6. Disentangled Mode-Specific Representations for Tensor Time Series via Contrastive Learning

    Feb 27, 2026Kohei Obata, Taichi Murayama, Zheng Chen +2Time Series Representation LearningDisentangled Representation Learning

    SANKEN, Osaka University, Osaka, Japan · Yokohama National University, Kanagawa, Japan

  7. A Statistical Approach for Modeling Irregular Multivariate Time Series with Missing Observations

    Feb 23, 2026Dingyi Nie, Yixing Wu, C. -C. Jay KuoIrregular Time-Series ModelingMultivariate Time Series Classification

    University of Southern California

  8. FRAME-C: A knowledge-augmented deep learning pipeline for classifying multi-electrode array electrophysiological signals

    May 18, 2025Nisal Ranasinghe, Dzung Do-Ha, Chathura J. Gamage +5Brain Disorder ClassificationMultivariate Time Series Classification

    AI, Optimization and Pattern Recognition Research Group, Dept. of Mechanical Eng., University of Melbourne, Australia · Illawarra Health and Medical Research Institute, School of Biological Sciences, Faculty of Science, Medicine and Health, University of Wollongong, Australia