mlsanityData to 14 Oct 2026

Assessing Distribution Shift in Human Activity Recognition for Domain Generalization

arXiv 2606.24781 ↗First posted 23 Jun 2026cs.AI cs.HCcited by 0

Not on the map. Why?

Abstract (arXiv)

While the field of Human Activity Recognition (HAR) continues to draw interest from researchers and advance in important ways, some key challenges remain. One of the most difficult aspects of building HAR models that show good performance in real-world settings is dealing with data diversity from device and sensor heterogeneity, and contextual changes that are intrinsic to real-world applications. While data diversity in HAR has been well-acknowledged in the literature, there remains a gap in understanding the effect of various types of distribution shifts on HAR models and the domain generalization problem that arises. Towards that end, this paper systematically evaluates 4 different types of distribution shifts, including variations in device type, sensor placement, sampling rate, and user behavior. Quantifying their effects, we illustrate that diversity shifts predominantly define all types of shifts, indicating the existence of unique features that are not shared across different domains. We then introduce a uniform HAR-based distribution shift benchmarks and conduct a comprehensive evaluation of up to 28 domain generalization methods. Our analysis exposes the limitations of current domain generalization algorithms in achieving model generalizability, marginally outperforming the empirical risk minimization baseline. This work represents the first systematic exploration of domain generalization and adaptation concerning specific distribution shifts in sensor-based HAR, offering an open-source benchmark platform and datasets to spur further research.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 32 of its 80 references that are in the map. First posted on arXiv 23 Jun 2026.

Builds on

32 of 80 references.

References · 32

  1. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  2. 2015Domain-Adversarial Training of Neural Networkscited by 3,044
  3. 2019Invariant Risk Minimizationcited by 1,854
  4. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,375
  5. 2019Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalizationcited by 1,247
  6. 2016Deep CORAL: Correlation Alignment for Deep Domain Adaptationcited by 1,200
  7. 2020In Search of Lost Domain Generalizationcited by 1,054
  8. 2021Generalizing to Unseen Domains: A Survey on Domain Generalizationcited by 766
  9. 2017Learning to Generalize: Meta-Learning for Domain Generalizationcited by 763
  10. 2020Out-of-Distribution Generalization via Risk Extrapolation (REx)cited by 742
  11. 2020Gradient Starvation: A Learning Proclivity in Neural Networkscited by 267
  12. 2021Gradient Matching for Domain Generalizationcited by 258
20 more
  1. 2021Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalizationcited by 220
  2. 2018Cross-Domain Weakly-Supervised Object Detection through Progressive Domain Adaptationcited by 218
  3. 2015Simultaneous Deep Transfer Across Domains and Taskscited by 206
  4. 2021SelfReg: Self-supervised Contrastive Regularization for Domain Generalizationcited by 189
  5. 2020Improve Unsupervised Domain Adaptation with Mixup Trainingcited by 185
  6. 2020Adaptive Risk Minimization: Learning to Adapt to Domain Shiftcited by 183
  7. 2017Domain Generalization by Marginal Transfer Learningcited by 180
  8. 2020Learning explanations that are hard to varycited by 132
  9. 2020Domain Generalization for Medical Imaging Classification with Linear-Dependency Regularizationcited by 123
  10. 2020Learning from Extrinsic and Intrinsic Supervisions for Domain Generalizationcited by 108
  11. 2021Model-Based Domain Generalizationcited by 104
  12. 2022Probable Domain Generalization via Quantile Risk Minimizationcited by 66
  13. 2020When is invariance useful in an Out-of-Distribution Generalization problem ?cited by 52
  14. 2021SAND-mask: An Enhanced Gradient Masking Strategy for the Discovery of Invariances in Domain Generalizationcited by 50
  15. 2021Quantifying and Improving Transferability in Domain Generalizationcited by 35
  16. 2022Invariant Causal Mechanisms through Distribution Matchingcited by 31
  17. 2022Semantic-Discriminative Mixup for Generalizable Sensor-based Cross-domain Activity Recognitioncited by 29
  18. 2018Deep Transfer Learning for Cross-domain Activity Recognitioncited by 25
  19. 2021Learning Representations that Support Robust Transfer of Predictorscited by 17
  20. 2020Using Convolutional Variational Autoencoders to Predict Post-Trauma Health Outcomes from Actigraphy Datacited by 2

Built on by

Not cited yet.