mlsanityData to 14 Oct 2026

Out-of-Distribution Representation Learning for Time Series Classification

arXiv 2209.07027 ↗First posted 15 Sep 2022cs.LG cs.AIcited by 31

Not on the map. Why?

Abstract (arXiv)

Time series classification is an important problem in real world. Due to its non-stationary property that the distribution changes over time, it remains challenging to build models for generalization to unseen distributions. In this paper, we propose to view the time series classification problem from the distribution perspective. We argue that the temporal complexity attributes to the unknown latent distributions within. To this end, we propose DIVERSIFY to learn generalized representations for time series classification. DIVERSIFY takes an iterative process: it first obtains the worst-case distribution scenario via adversarial training, then matches the distributions of the obtained sub-domains. We also present some theoretical insights. We conduct experiments on gesture recognition, speech commands recognition, wearable stress and affect detection, and sensor-based human activity recognition with a total of seven datasets in different settings. Results demonstrate that DIVERSIFY significantly outperforms other baselines and effectively characterizes the latent distributions by qualitative and quantitative analysis. Code is available at: https://github.com/microsoft/robustlearn.

Key facts. Cited by 31 papers in the map (data through 14 Oct 2026). Builds on 39 of its 67 references that are in the map. First posted on arXiv 15 Sep 2022.

Builds on

39 of 67 references.

References · 39

  1. 2017Attention Is All You Needcited by 60,361
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  4. 2015Domain-Adversarial Training of Neural Networkscited by 3,044
  5. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,375
  6. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,247
  7. 2016Deep CORAL: Correlation Alignment for Deep Domain Adaptationcited by 1,200
  8. 2020In Search of Lost Domain Generalizationcited by 1,054
  9. 2018Deep Clustering for Unsupervised Learning of Visual Featurescited by 1,028
  10. 2021Generalizing to Unseen Domains: A Survey on Domain Generalizationcited by 766
  11. 2018Speech Commands: A Dataset for Limited-Vocabulary Speech Recognitioncited by 752
  12. 2020Out-of-Distribution Generalization via Risk Extrapolation (REx)cited by 742
27 more
  1. 2013Domain Generalization via Invariant Feature Representationcited by 623
  2. 2019Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecastingcited by 607
  3. 2018Deep learning for time series classification: a reviewcited by 576
  4. 2020Neural Controlled Differential Equations for Irregular Time Seriescited by 506
  5. 2020Self-Challenging Improves Cross-Domain Generalizationcited by 426
  6. 2020MINIROCKET: A Very Fast (Almost) Deterministic Transform for Time Series Classificationcited by 190
  7. 2021Fishr: Invariant Gradient Variances for Out-of-Distribution Generalizationcited by 186
  8. 2019Domain Generalization Using a Mixture of Multiple Latent Domainscited by 170
  9. 2021Learning to Diversify for Single Domain Generalizationcited by 170
  10. 2017Deep Learning for Sensor-based Activity Recognition: A Surveycited by 133
  11. 2020Learning explanations that are hard to varycited by 132
  12. 2021AdaRNN: Adaptive Learning and Forecasting of Time Seriescited by 121
  13. 2019Domain Agnostic Learning with Disentangled Representationscited by 108
  14. 2021Progressive Domain Expansion Network for Single Domain Generalizationcited by 97
  15. 2021Towards Principled Disentanglement for Domain Generalizationcited by 91
  16. 2021Adversarially Adaptive Normalization for Single Domain Generalizationcited by 85
  17. 2021Delving Deep into the Generalization of Vision Transformers under Distribution Shiftscited by 85
  18. 2020Multi-Source Deep Domain Adaptation with Weak Supervision for Time-Series Sensor Datacited by 82
  19. 2020Continuously Indexed Domain Adaptationcited by 75
  20. 2021Domain Adversarial Neural Networks for Domain Generalization: When It Works and How to Improvecited by 46
  21. 2014Highly comparative feature-based time-series classificationcited by 40
  22. 2022TACTiS: Transformer-Attentional Copulas for Time Seriescited by 38
  23. 2020MatchboxNet: 1D Time-Channel Separable Convolutional Neural Network Architecture for Speech Commands Recognitioncited by 36
  24. 2022Domain-invariant Feature Exploration for Domain Generalizationcited by 34
  25. 2022Provable Domain Generalization via Invariant-Feature Subspace Recoverycited by 32
  26. 2021Robust Domain-Free Domain Generalization with Class-aware Alignmentcited by 6
  27. 2021Improving Model Generalization by Agreement of Learned Representations from Data Augmentationcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 4; 2024: 13; 2025: 11; 2026: 3.

31 papers.

Sort:

By area:

  1. 2024UniTS: A Unified Multi-Task Time Series Modelcited by 94
  2. 2023SimMMDG: A Simple and Effective Framework for Multi-modal Domain Generalizationcited by 48
  3. 2024Time-Series Forecasting for Out-of-Distribution Generalization Using Invariant Learningcited by 27
  4. 2024Towards Multimodal Open-Set Domain Generalization and Adaptation through Self-supervisioncited by 16
  5. 2025Out-of-Distribution Generalization in Time Series: A Surveycited by 16
  6. 2025SciTS: Scientific Time Series Understanding and Generation with LLMscited by 15
  7. 2023Unsupervised Representation Learning for Time Series: A Reviewcited by 14
  8. 2024Tri-Level Navigator: LLM-Empowered Tri-Level Learning for Time Series OOD Generalizationcited by 12
  9. 2024Evaluating the Generalization Ability of Spatiotemporal Model in Urban Scenariocited by 7
  10. 2023Continuous Invariance Learningcited by 6
  11. 2025Simulations of Common Unsupervised Domain Adaptation Algorithms for Image Classificationcited by 6
  12. 2025StruSR: Structure-Aware Symbolic Regression with Physics-Informed Taylor Guidancecited by 6
19 more
  1. 2024Phase-driven Domain Generalizable Learning for Nonstationary Time Seriescited by 5
  2. 2023Target-agnostic Source-free Domain Adaptation for Regression Taskscited by 3
  3. 2024Deep Generative Domain Adaptation with Temporal Attention for Cross-User Activity Recognitioncited by 3
  4. 2024A Scene-aware Models Adaptation Scheme for Cross-scene Online Inference on Mobile Devicescited by 3
  5. 2024FredNormer: Frequency Domain Normalization for Non-stationary Time Series Forecastingcited by 3
  6. 2025HAR-DoReMi: Optimizing Data Mixture for Self-Supervised Human Activity Recognition Across Heterogeneous IMU Datasetscited by 2
  7. 2025Continuous Domain Generalizationcited by 2
  8. 2025SSAS: Cross-subject EEG-based Emotion Recognition through Source Selection with Adversarial Strategycited by 2
  9. 2025Prism: Mining Task-aware Domains in Non-i.i.d. IMU Data for Flexible User Perceptioncited by 1
  10. 2025Tackling Time-Series Forecasting Generalization via Mitigating Concept Driftcited by 1
  11. 2024Cross-user activity recognition using deep domain adaptation with temporal relation informationcited by 0
  12. 2024Elevator, Escalator, or Neither? Classifying Conveyor State Using Smartphone under Arbitrary Pedestrian Behaviorcited by 0
  13. 2024Adversarial Domain Adaptation for Cross-user Activity Recognition Using Diffusion-based Noise-centred Learningcited by 0
  14. 2024Learning Latent Spaces for Domain Generalization in Time Series Forecastingcited by 0
  15. 2025ERIS: An Energy-Guided Feature Disentanglement Framework for Out-of-Distribution Time Series Classificationcited by 0
  16. 2025Out of Distribution Detection for Efficient Continual Learning in Quality Prediction for Arc Weldingcited by 0
  17. 2026ADAPTOOD: Uncertainty-Aware Fine-Tuning for Out-of-Distribution ECG Time Series Modelscited by 0
  18. 2026Domain Generalization under Sampling Pattern Shifts in Irregular Time Seriescited by 0
  19. 2026W-JEPA: Wavelet Joint-Embedding Predictive Architecture for Wearable Activity Recognitioncited by 0