The DOI matches its Crossref record; 67 of the 67 references it cites were found in a registry or an index; 8 of the 44 sources its claims cite were checked for support.
Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches
Open accessAbstract
Human activity recognition (HAR) using wearable sensors has advanced through various machine learning paradigms, each with inherent trade-offs between performance and labeling requirements. While fully supervised techniques achieve high accuracy, they demand extensive labeled datasets that are costly to obtain. Conversely, unsupervised methods eliminate labeling needs but often deliver suboptimal performance. This paper presents a comprehensive investigation across the supervision spectrum for wearable-based HAR, with particular focus on novel approaches that minimize labeling requirements while maintaining competitive accuracy. We develop and empirically compare: (1) traditional fully supervised learning, (2) basic unsupervised learning, (3) a weakly supervised learning approach with constraints, (4) a multi-task learning approach with knowledge sharing, (5) a self-supervised approach based on domain expertise, and (6) a novel weakly self-supervised learning framework that leverages domain knowledge and minimal labeled data. Experiments across benchmark datasets demonstrate that: (i) our weakly supervised methods achieve performance comparable to fully supervised approaches while significantly reducing supervision requirements; (ii) the proposed multi-task framework enhances performance through knowledge sharing between related tasks; (iii) our weakly self-supervised approach demonstrates remarkable efficiency with just 10% of labeled data. These results not only highlight the complementary strengths of different learning paradigms, offering insights into tailoring HAR solutions based on the availability of labeled data, but also establish that our novel weakly self-supervised framework offers a promising solution for practical HAR applications where labeled data are limited.
Overview
CoChat Summary
TL;DR
The paper compares supervised, unsupervised, weakly supervised, self-supervised, and hybrid weakly self-supervised methods for wearable-sensor human activity recognition (HAR) to reduce label dependence. Across three benchmark datasets, the weakly supervised and hybrid methods are reported to approach or exceed fully supervised performance while using far fewer labels.
Key findings
- The weakly supervised single-task method achieved 94.36% accuracy on UCI-Smartphone, 98.56% on PAMAP2, and 97.96% on REALDISP, substantially outperforming the unsupervised baseline (63.69%, 77.06%, 64.01%).
- The weakly supervised multi-task method achieved 98.85%, 98.93%, and 98.22% accuracy on the three datasets, and on UCI-Smartphone it exceeded the supervised TCN baseline (98.85% vs. 98.66%).
- The self-supervised method outperformed the standard autoencoder on all datasets: 74.01% vs. 63.69% on UCI-Smartphone, 85.43% vs. 77.06% on PAMAP2, and 68.12% vs. 64.01% on REALDISP.
- The weakly self-supervised method improved further with more labels: with 1% labeled data it reached 80.15%, 91.28%, and 75.15%, and with 10% labeled data it reached 93.52%, 99.04%, and 92.23%; on PAMAP2, 99.04% exceeded the fully supervised TCN baseline of 95.27%.
Methods
- The study systematically compares six paradigms: fully supervised learning, unsupervised autoencoders on handcrafted statistical features, weakly supervised Siamese/contrastive learning with pairwise similarity constraints, weakly supervised multi-task learning with an added person-identification task, self-supervised learning with temporal and feature consistency losses, and a two-stage weakly self-supervised framework combining self-supervised pretraining with weak supervision.
- Evaluation was conducted on three benchmark HAR datasets: UCI-Smartphone (30 participants, 12 activities), PAMAP2 (9 participants, 12 activities), and REALDISP (17 participants, 33 activities).
- The paper reports using accuracy and F1-score as metrics, with stratified splits and k-means clustering on learned latent representations for the non-classification approaches.
Limitations
- The paper notes that fully supervised methods require extensive labeled data, while unsupervised methods often perform worse due to lack of guidance; this is stated as a general trade-off in the paper.
- The authors state that temporal-consistency self-supervision can struggle on activity transitions such as sit-to-stand or stand-to-walk, where signal patterns are ambiguous; this may limit effectiveness.
- Evident from the text: the evaluation is restricted to three benchmark datasets and to ideal-placement settings for REALDISP, so the reported results may not fully cover broader real-world deployment variability.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures









Verification
Claims the paper makes
5 Valid · 0 Not valid · 18 Not checked · 94 Own results and interpretations
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“HAR and person identification exhibit a strong connection”
§ Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches > 1. Introduction
What the sources say
- Not checked yetcites [3] Ruder, S. (2017). An Overview of Multi-Task Learning in Deep Neural Networks. arXiv.
- Supportedcites [4] Sheng, 2020, Weakly supervised multi-task representation learning for human activity analysis using wearables, Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. source
The Introduction states that human activity recognition (HAR) and person identification are closely related. The Ablation Studies section says the results confirm that relationship, directly supporting the paraphrase. (LLM, conf=0.94)
- Could not verifycites [5] Saeed, 2019, Multi-task Self-Supervised Learning for Human Activity Detection, Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. sourceExperimental — accuracy not yet measured
The paper’s Introduction (p. 1) and Evaluation (p. 8) discuss human activity recognition and learned sensor representations, but do not address a connection to person identification. Searches for person identification, user identification, and subject identity found no passage making that connection. (LLM, conf=0.73)
boostedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“the proposed multi-task framework enhances performance through knowledge sharing between related tasks”
§ Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches > Abstract Human activity recognition (HAR) using wearable sensors has advanced through various machine learning paradigms, each with inherent trade-offs between performance and labeling requirements. While fully supervised techniques achieve high accuracy, they demand extensive labeled datasets that are costly to obtain. Conversely, unsupervised methods eliminate labeling needs but often deliver suboptimal performance. This paper presents a comprehensive investigation across the supervision spectrum for wearable-based HAR, with particular focus on novel approaches that minimize labeling requirements while maintaining competitive accuracy. We develop and empirically compare: (1) traditional fully supervised learning, (2) basic unsupervised learning, (3) a weakly supervised learning approach with constraints, (4) a multi-task learning approach with knowledge sharing, (5) a self-supervised approach based on domain expertise, and (6) a novel weakly self-supervised learning framework that leverages domain knowledge and minimal labeled data. Experiments across benchmark datasets demonstrate that: (i) our weakly supervised methods achieve performance comparable to fully supervised approaches while significantly reducing supervision requirements; (ii) the proposed multi-task framework enhances performance through knowledge sharing between related tasks; (iii) our weakly self-supervised approach demonstrates remarkable efficiency with just 10% of labeled data. These results not only highlight the complementary strengths of different learning paradigms, offering insights into tailoring HAR solutions based on the availability of labeled data, but also establish that our novel weakly self-supervised framework offers a promising solution for practical HAR applications where labeled data are limited.
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“our weakly self-supervised approach demonstrates remarkable efficiency with just 10% of labeled data”
§ Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches > Abstract Human activity recognition (HAR) using wearable sensors has advanced through various machine learning paradigms, each with inherent trade-offs between performance and labeling requirements. While fully supervised techniques achieve high accuracy, they demand extensive labeled datasets that are costly to obtain. Conversely, unsupervised methods eliminate labeling needs but often deliver suboptimal performance. This paper presents a comprehensive investigation across the supervision spectrum for wearable-based HAR, with particular focus on novel approaches that minimize labeling requirements while maintaining competitive accuracy. We develop and empirically compare: (1) traditional fully supervised learning, (2) basic unsupervised learning, (3) a weakly supervised learning approach with constraints, (4) a multi-task learning approach with knowledge sharing, (5) a self-supervised approach based on domain expertise, and (6) a novel weakly self-supervised learning framework that leverages domain knowledge and minimal labeled data. Experiments across benchmark datasets demonstrate that: (i) our weakly supervised methods achieve performance comparable to fully supervised approaches while significantly reducing supervision requirements; (ii) the proposed multi-task framework enhances performance through knowledge sharing between related tasks; (iii) our weakly self-supervised approach demonstrates remarkable efficiency with just 10% of labeled data. These results not only highlight the complementary strengths of different learning paradigms, offering insights into tailoring HAR solutions based on the availability of labeled data, but also establish that our novel weakly self-supervised framework offers a promising solution for practical HAR applications where labeled data are limited.
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“we conducted experiments with extremely limited label availability (1%, 5%, and 10% of labeled data), showing how our combined framework progressively improves performance as more labels become available”
§ Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches > 1. Introduction
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Its performance demonstrates a clear trend: as the amount of labeled data increases, so does the model’s effectiveness.”
§ Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches > 4. Evaluation and Experiments > 4.2. Results and Analysis
boostedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
46 more claims
Create a free account to see all 117 claims and to verify the 36 references still open.
- Probably not a substantial claim. 34
- It may state two separate claims as one. 41
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches > 4. Evaluation and Experiments > 4.2. Results and Analysis
“While it does not consistently outperform all supervised methods, it achieves results comparable to or better than several fully supervised techniques despite not requiring explicitly labeled data during training.”
§ Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches > 4. Evaluation and Experiments > 4.3. Visualization
“As shown in Figure 8a,b, the weakly supervised single-task and multi-task approaches produce the most distinct clusters, benefiting from the use of all labeled data.”
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 2×
“Each selected technique addresses specific challenges in HAR: autoencoders [36] for unsupervised representation learning, ResNet [37] for handling deep architectures without degradation, Siamese networks [38] for similarity-based learning in weakly supervised settings, temporal convolutional…”
(2008) · Journal of Machine Learning Research · Cited by 33,051
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.81 · Cited by 32,684
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 1×
“Each convolutional layer is followed by batch normalization (BN) [49], which stabilizes optimization by reparameterizing the problem into a more tractable form.”
(2010) · International Conference on Machine Learning · Cited by 13,995
Scaling Laws for Neural Language Models
Kaplan, J., McCandlish, S., Henighan, T., Brown, T. B., Chess, B., Child, R., Gray, S., Radford, A., Wu, J., & Amodei, D. (2020) · arXiv · doi:10.48550/arxiv.2001.08361 · Cited by 9,550 · cited 1×
“This characteristic holds significant value, given that the magnitude of the dataset plays a pivotal role in enhancing the real-world performance of models in downstream tasks, as highlighted by studies such as [27].”
Extracting and composing robust features with denoising autoencoders
Vincent, P., Larochelle, H., Bengio, Y., & Manzagol, P.-A. (2008) · Proceedings of the 25th international conference on Machine learning - ICML '08 · doi:10.1145/1390156.1390294 · Cited by 7,436 · cited 1×
“For instance, Vincent et al. [41] used denoising autoencoders for robust feature learning, Kingma and Welling [42] introduced variational autoencoders for generative modeling, and Sakurada and Yairi [43] employed autoencoders for anomaly detection in time-series data.”
Dimensionality Reduction by Learning an Invariant Mapping
Hadsell, R., Chopra, S., & LeCun, Y. (2006) · 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2 (CVPR'06) · doi:10.1109/cvpr.2006.100 · Cited by 5,336 · cited 1×
“…deep architectures without degradation, Siamese networks [38] for similarity-based learning in weakly supervised settings, temporal convolutional networks (TCNs) [39] for effective temporal pattern extraction, and contrastive loss functions [40] for implementing pairwise constraint-based learning.”
Learning a Similarity Metric Discriminatively, with Application to Face Verification
Chopra, S., Hadsell, R., & LeCun, Y. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.202 · Cited by 4,036 · cited 2×
“We first developed a weakly supervised approach inspired by previous research in computer vision [1,2].”
InfluentialSiamese Neural Networks for One-shot Image Recognition
Koch, G. E., Zemel, R. S., & Salakhutdinov, R. (2015) · Cited by 3,740
Research works printed but not linked: 8
- Ruder, S. (2017). An Overview of Multi-Task Learning in Deep Neural Networks. arXiv.
- Chen, R., Luo, H., Zhao, F., Meng, X., Xie, Z., and Zhu, Y. (2021). Modeling Accurate Human Activity Recognition for Embedded Devices Using Multi-level Distillation. arXiv.
- Simonyan, K., and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv.
- McInnes, L., Healy, J., and Melville, J. (2020). UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. arXiv.
- Mikolov, T., Grave, E., Bojanowski, P., Puhrsch, C., and Joulin, A. (2017). Advances in pre-training distributed word representations. arXiv.
- Kingma, D.P., and Welling, M. (2022). Auto-Encoding Variational Bayes. arXiv.
- Chen, T., Kornblith, S., Norouzi, M., and Hinton, G.E. (2020). A Simple Framework for Contrastive Learning of Visual Representations. arXiv.
- Aljalbout, E., Golkov, V., Siddiqui, Y., and Cremers, D. (2018). Clustering with Deep Learning: Taxonomy and New Methods. arXiv.
Cited by
Self-Supervised Gait Event Detection from Smartphone IMUs for Human Performance and Sports Medicine
TUDOR, A. C., Mănescu, D. C., & Mănescu, D. C. (2025) · Applied Sciences · doi:10.3390/app152211974 · Cited by 18
Kong, M., Zhao, X., Li, Q., Yu, Y., & Zhang, C. (2025) · Advanced Intelligent Systems · doi:10.1002/aisy.202501136 · Cited by 2
Liu, X., & Wang, G. (2026) · Mathematics · doi:10.3390/math14111932
Biomechanics-Informed Masked Autoencoder With Anchor Alignment for Robust Human Activity Recognition
Guo, T., Hou, F., Pang, Y., & Wang, Z. (2026) · IEEE Access · doi:10.1109/access.2026.3713236
“The present convergence of advanced wearable sensor capabilities, sophisticated self-supervised learning (SSL) paradigms, and a deepened understanding of human biomechanics presents a timely and unique opportunity to address these long-standing challenges [6], [7].”
Reducing Labeled Data Dependency with Self-Supervised Learning: Advances and applications
Tyagi, H., Bhardwaj, M. K., Bipin, Tyagi, S., Singh, P., & Himanshu (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-30005-8_20
Metadata
- Authors
- Sheng, T., & Huber, M.
- Affiliations
- The University of Texas at Arlington (US)
- Corresponding author
- Sheng, Taoran
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2025-06-28
- Volume
- 25
- Issue
- 13
- Pages
- 4032
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC12252065
- OpenAlex
- W4411802278
- Cited by
- 5
- References
- 63
- Citations by year
- 2026: 4 · 2025: 1
- FWCI
- 1.76
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Signal Processing › Time Series Analysis and Forecasting
- Keywords
- human activity recognition, wearable sensors, weakly supervised learning, multi-task learning, knowledge sharing, domain knowledge, limited labeled data, unsupervised learning, supervised learning
- MeSH terms
- Machine Learning, Supervised Machine Learning, Unsupervised Machine Learning, Wearable Electronic Devices, Algorithms, Humans, Human Activities
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches. https://app.cochat.ai/papers/10.3390/s25134032
- The paper
- Sheng, T., & Huber, M. (2025). Reducing Label Dependency in Human Activity Recognition with Wearables: From Supervised Learning to Novel Weakly Self-Supervised Approaches. Sensors. https://doi.org/10.3390/s25134032
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.