Self-Supervised Time Series Representation Learning

Overview

Self-supervised time series representation learning studies how to learn useful features from sequences without relying on extensive labels, which are often costly or unavailable. Common approaches include contrastive learning between related views, masked reconstruction, generative prediction, and joint-embedding predictive architectures (JEPAs), implemented with convolutional networks, Transformers, or state-space models. These representations are tested across tasks such as activity recognition, forecasting, anomaly detection, and clinical analysis of ECG, EEG, and wearable signals, often using frozen linear probes, fine-tuning, and cross-dataset transfer. A central open question is which objectives and inductive biases help each task: gains vary across datasets and architectures, while invariance to nuisance variation can also erase meaningful changes over time.

Latest papers

CardsList

164 papers20 papers in the last four weeks, down 23% on the four weeks before. 0.2% of all new papers.

  1. Breaking feature collapse in self-supervised time series encoders

    Nov 16, 2026Özgün Turgut, Daniel RueckertSelf-Supervised Time Series Representation LearningRepresentation Collapse

  2. HeteroWear: Sensor-Aware Self-Supervised Learning for Heterogeneous Wearable Activity Recognition

    Nov 13, 2026Yasaman Mohammadpour, Pouya ShaeriWearable SensingHuman Activity Recognition

  3. SLICES: A Benchmark and Evaluation Framework for Self-Supervised Learning on Sparse ICU Time Series

    Nov 13, 2026Hannes Ill, Dimitrios Proios, Leo Anthony Celi +1Self-Supervised Time Series Representation LearningClinical Outcome Prediction

  4. Self-Supervised Attention-Based Virtual Metrology for CMP MRR Prediction

    Nov 12, 2026Jonathan Adam Rico, Nagarajan Raghavan, Senthilnath JayaveluManufacturingSemiconductor Manufacturing

  5. Market-Conditioned JEPA: Learning from Unpredictable Market Shocks

    Nov 10, 2026Yuqi Wang, Fengyuan Liu, Mukai Li +3Financial Time SeriesQuantitative Finance

  6. Cross-Domain Joint-Embedding Pretraining For Heterogeneous Industrial Time Series

    Nov 2, 2026Jonathan Adam Rico, Nagarajan Raghavan, Senthilnath JayaveluSelf-Supervised Time Series Representation LearningSelf-Supervised Pre-Training

  7. SensOFormer: Foundation Models for Particle Accelerators

    Nov 1, 2026Mahindra Rautela, Alexander ScheinkerSelf-Supervised Time Series Representation LearningTime-Series Foundation Models

  8. HieraNTP: Structured Risk Representations for Time-Series Anomaly Detection via Hierarchical Next-Token Prediction

    Oct 31, 2026Yujie Li, Zezhi Shao, Boyu Diao +4Time-Series Anomaly DetectionAnomaly Detection

  9. SAR-TS-BERT: Physics-Guided Self-Supervised Pre-Training With Polarimetric Features for SAR Time-Series Crop Classification

    Oct 30, 2026Jieqiu Wang, Yuhan Zhang, Bo Zheng +1Agricultural Remote SensingSynthetic Aperture Radar

  10. Contrastive Astromer: Two-Stage Contrastive Representation Learning for Astronomical Light Curve Classification

    Oct 30, 2026Justin Minggan Wei, Pavlos Protopapas, Guillermo Cabrera-VivesContrastive LearningSelf-Supervised Time Series Representation Learning

  11. Are Seismic Foundation Models More Than Glorified Feature Extractors?

    Oct 27, 2026Laura Laurenti, David Liu, Brian Kulis +2Time-Series Foundation ModelsSelf-Supervised Time Series Representation Learning

  12. Learning from Points and Patches: A Pre-trained Hybrid Attention Transformer for Time Series Analysis

    Oct 27, 2026Peiwang Tang, Cheng HE, Yupeng Ye +3Self-Supervised Time Series Representation LearningTime Series Transformers

  13. TransECG-SSL: Hybrid Self-Supervised Representation Learning for ECG Arrhythmia Classification

    Oct 26, 2026Ahtisham Ayyub, Fasih Haider, Muhammad Rehan Usman +3Electrocardiogram Representation LearningElectrocardiogram Classification

  14. YingLan: A Multimodal Ocean Foundation Model for Unified Representation Learning from Heterogeneous Observations

    Oct 21, 2026Sen Jia, Siqi Wei, Junyu Wang +8Multimodal Foundation ModelsSpatiotemporal Forecasting

  15. Turning TS2Vec into a Foundation Model for Time Series Classification

    Oct 20, 2026Vasilii Feofanov, Zhenghan Tai, Andrei Ermakov +1Time Series ClassificationTime-Series Foundation Models

  16. Representing short-term movement to reveal animal identity and learning progress

    Oct 19, 2026Ping Zhang, Quang-Huy Nguyen, Yingtie Lei +5Animal Behavior AnalysisTrajectory Representation Learning

  17. Global Time-Aware Conditioning (GTC): A Mechanism for Time-Aware Pediatric Sleep Representation Learning

    Oct 19, 2026Saurav Raj Pandey, Harlin LeeTemporal Representation LearningSelf-Supervised Time Series Representation Learning

  18. OliO: ODE-based Linear Transition Operator for Self-Supervised Time Series Forecasting

    Oct 16, 2026Kwangryeol Park, Seulki LeeTime Series ForecastingSelf-Supervised Time Series Representation Learning

  19. Local Temporal Contrasts on Human Motion Data Induce a Latent Geometry

    Oct 15, 2026Lourenço A. Rodrigues, Felix PutzeTemporal Representation LearningSelf-Supervised Time Series Representation Learning

  20. TEX-MEX: Similarity-Guided Temporal Contrastive Learning for Time Series

    Oct 14, 2026Aditya Radhakrishnan, Maxwell A. Xu, Wanting Mao +3Self-Supervised Time Series Representation LearningContrastive Learning

  21. W-JEPA: Wavelet Joint-Embedding Predictive Architecture for Wearable Activity Recognition

    Oct 14, 2026Yogesh Prabhu, Loki Natarajan, Jingjing ZouHuman Activity RecognitionRepresentation Learning

  22. On Sleep-Wake Disorder Identification from Polysomnographic Representations

    Oct 12, 2026Alvise Dei Rossi, Luigi Fiorillo, Giuliana Monachino +6Self-Supervised Pre-TrainingSelf-Supervised Time Series Representation Learning

  23. Variable Input Masked Autoencoders for Wrist-Worn Accelerometry

    Oct 12, 2026Harish Haresamudram, Wanting Mao, Aditya Radhakrishnan +2Self-Supervised Time Series Representation LearningTime-Series Foundation Models

  24. KITE: A Knowledge-Injected Temporal Encoder for Time-Series Event Forecasting

    Oct 8, 2026Jonas Petersen, Pratik Jawahar, Riccardo Maggioni +6Rare-Event PredictionTime Series Forecasting