Self-supervised learning for human activity recognition using 700,000 person-days of wearable data
Open accessThe DOI matches its Crossref record; 73 of the 73 references it cites were found in a registry or an index; 17 of the 20 sources its claims cite were checked for support.
Abstract
A large-scale study leveraging self-supervised learning on 700,000 person-days of unlabeled accelerometer data from the UK Biobank to build generalized activity recognition models that outperform supervised baselines across eight benchmark datasets with a median relative F1 improvement of 24.4%.
Overview
CoChat Summary
TL;DR
This paper applies multi-task self-supervised learning to 700,000 person-days of unlabelled UK Biobank wrist accelerometer data, then transfers the pre-trained model to eight human activity recognition benchmarks. The resulting models consistently outperform random forest and from-scratch deep learning baselines, with F1 relative improvements of 2.5–130.9% (median 24.4%).
Key findings
- Pre-trained models outperformed both random forest and ResNet trained from scratch on all eight downstream datasets; the largest gains were on small datasets, e.g. MJFF-LR F1 0.755 ± 0.109 vs 0.327 ± 0.103 from scratch (130.9% improvement), ADL 0.829 ± 0.101 vs 0.414 ± 0.179 (100.0%), and Opportunity 0.595 ± 0.085 vs 0.383 ± 0.124 (55.4%).
- The smallest relative improvement was on the largest labelled dataset, Capture-24: F1 0.726 ± 0.093 with self-supervised fine-tuning of all layers vs 0.708 ± 0.094 from scratch (2.5%).
- Fine-tuning all layers was better than fine-tuning only the fully connected layers after the Conv layers, and self-supervised pre-training on UK Biobank generally outperformed supervised pre-training on Rowlands or Capture-24 in the transfer-learning comparisons.
- Downstream performance increased roughly linearly with the number of unlabelled pre-training subjects on a log scale, and reducing data per subject from 100% to 25% did not degrade F1 by more than 10%.
Methods
- Multi-task self-supervised pre-training used unlabelled UK Biobank wrist accelerometer data (~100,000 participants, 7 days, >700,000 person-days), resampled to 30 Hz with 10-second windows. The network was an 18-layer 1D ResNet-V2 with 10M parameters and a 1024-dimensional feature vector; pretext tasks were arrow of time, permutation, and time warping, with weighted sampling by signal standard deviation. Training used Adam with learning rate 1e-3, four Tesla V100-SXM2 GPUs, and about 420 GPU hours for ~20 epochs.
- Evaluation used eight external labelled datasets varying in size (600–600,000 samples), activity classes (4–18), devices (5 brands), placements (4 configurations), and populations. Models were fine-tuned either all layers or only FC layers and compared against random forest and from-scratch ResNet; held-one-subject-out cross-validation was used for datasets with <10 subjects, and five-fold subject-wise cross-validation for datasets with ≥10 subjects, with a 7:1:2 train/validation/test split and early stopping patience of five.
- Ablations varied labelled downstream subjects and unlabelled pre-training participants (100 to 100,000) and data per subject (0.25–1); representation analyses used UMAP projections and layer-wise relevance propagation.
Limitations
- The paper states that the pre-training data (UK Biobank) consists mostly of Caucasians from the UK, due to a current lack of raw accelerometer datasets from different regions of the world. It also notes future work should consider multimodal wearable data, more recent self-supervised approaches, and that Autoencoder and contrastive learning attempts did not yield high-quality representations on UK Biobank, suspected to be due to free-living vs lab-based data differences.
- Evident from the text: evaluation is limited to wrist-worn accelerometer data and eight benchmark datasets, with no external validation in non-UK or non-Caucasian populations. The paper also reports that many open benchmark datasets used lacked suitable licensing or consent information, and small datasets had activity classes removed if not performed by all subjects.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures




Verification
Claims the paper makes
8 Valid · 3 Not valid · 0 Not checked · 55 Own results and interpretations
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“The scale of UK Biobank is several orders of magnitude greater than the largest datasets used by the current state-of-the-art such as the Fenland data9 (n = 160,000).”
§ Self-supervised learning for human activity recognition using 700,000 person-days of wearable data > Discussion
What the sources say
- Disputedcites [9] Lindsay, T. et al. Descriptive epidemiology of physical activity energy expenditure in UK adults (the Fenland study). Int. J. Behav. Nutr. Phys. Activity 16, 1–13 (2019). sourceExperimental — accuracy not yet measured
The Methods (Study population) state that the Fenland Study enrolled 12,435 participants, not 160,000. The Results further report that 12,002 met the sensing criteria for this analysis. The paper does not provide a UK Biobank comparison, so it cannot establish the claimed difference in scale. (LLM, conf=0.90)
Checked by a reasoning model — may be wrong.
plainn = 160,000- 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.
- Disputed
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“We attempted to use Autoencoder and contrastive learning for pre-training. However, we could not obtain high-quality representation using the UK Biobank43.”
§ Self-supervised learning for human activity recognition using 700,000 person-days of wearable data > Discussion
What the sources say
- Could not verifycites [43] Tang, C. I. et al. SelfHAR: Improving Human Activity Recognition through Self-training with Unlabeled Data. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 5, 30 (2021). sourceExperimental — accuracy not yet measured
The Experimental Protocols (§4.1.8) describe the Fenland Study, and the Results (§5.1.1) name MobiAct and HHAR as unlabeled datasets; neither passage addresses UK Biobank. Searches for UK Biobank, autoencoder and contrastive pre-training found no passage reporting the claimed failure to obtain a high-quality representation. The citing authors describe their own attempt, which this source cannot verify. (LLM, conf=0.76)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
Every source the paper cites for this claim was checked and supports it.
In the paper
“Differences in experiment setup such as training rates, regularisation and data augmentation can lead to inconsistent results51.”
§ Self-supervised learning for human activity recognition using 700,000 person-days of wearable data > Methods > Evaluation—human activity recognition
What the sources say
- Supportedcites [51] Oliver, A., Odena, A., Raffel, C. A., Cubuk, E. D. & Goodfellow, I. J. Realistic evaluation of deep semi-supervised learning algorithms. In Proceedings of the 32nd International Conference on Neural Information Processing Systems (NIPS'18), Vol. 31, 3239–3250 (Curran Associates Inc., Red Hook, NY, USA, 2018). source
Section 2 'P.1 A Shared Implementation' (p. 2) states that although datasets have standardized, other experimental details vary significantly — parameter initialization, data preprocessing, data augmentation, regularization — and that the training procedure (optimizer, number of training steps, learning rate decay schedule) is not standardized, and that 'these differences prevent direct comparison between approaches'. The Introduction (p. 2) reinforces this: 'minor changes to hyperparameters, model structure, training, etc. can have an outsized impact on results', and Section 4.2 (p. 6) adds that different algorithms must be evaluated on the same model to avoid conflating comparison. The claim's examples (training rates, regularisation, data augmentation) map directly onto these passages; 'inconsistent results' paraphrases the paper's point that results become non-comparable/confounded. (LLM, conf=0.88)
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.
- Supported
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
“Furthermore, if the number of participants is fixed at 10,000 in pre-training, the data ratio included per subject did not significantly influence the downstream performance.”
§ Self-supervised learning for human activity recognition using 700,000 person-days of wearable data > Results > Ablation studies
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether 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
“Nonetheless, with self-supervised pre-training, the median relative F1 improvement was 18.4% when compared to the same network trained from scratch and 8.5% when compared to the random forest model.”
§ Self-supervised learning for human activity recognition using 700,000 person-days of wearable data > Results > Downstream performance—human activity recognition
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.
39 more claims
Create a free account to see all 66 claims and every source verdict.
- It may state two separate claims as one. 13
- Probably not a substantial claim. 14
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Owing to the relative ease of data collection using wearables, there are now large repositories of datasets, e.g., UK Biobank [212], Capture-24 [21,198], and Fenland [145], but crucially, without annotations.”
From Past, Present, and Future of Sensor-based Human Activity Recognition Using Wearables: A Surveying Tutorial on a Still Challenging Task (2025) · cites this paper as [212]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“However, the gap between HAR-FM and the other foundation model candidates like UniMTS and YUAN-SSL suggests that incorporating larger datasets (evidenced by the gaps between baseline YUAN-SSL (capture24) and YUAN-SSL w10 ) and more fine-grained activity descriptions (evidenced by the significant performance of UniMTS) could lead to significant performance improvements.”
From Towards Customizable Foundation Models for Human Activity Recognition with Wearable Devices (2025) · cites this paper as [62]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Secondly, in the zero-and few-shot experiments, Yuan et al. [62] and Zhang et al. [65] demonstrates either larger volume or more precise annotations could contribute to the model performance”
From Towards Customizable Foundation Models for Human Activity Recognition with Wearable Devices (2025) · cites this paper as [62]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Its superior performance against HAR-FM and YUAN-SSL [62] that are pretrained with data from wrist only, suggests the benefits in integrating all placement-wise models.”
From Towards Customizable Foundation Models for Human Activity Recognition with Wearable Devices (2025) · cites this paper as [62]
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
§ Self-supervised learning for human activity recognition using 700,000 person-days of wearable data > Methods > Evaluation—human activity recognition
“In terms of unlabelled data, we experimented with pre-training that had 100 to 100,0000 participants with one order of magnitude increment.”
§ Self-supervised learning for human activity recognition using 700,000 person-days of wearable data > Methods > Datasets
“Our multi-task self-supervised training relied on the unlabelledUK Biobank dataset, which contains roughly 700,000 person-days of free-living activity data (>100,000 participants, 7 days of wear).”
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2018) · arXiv · doi:10.48550/arxiv.1810.04805 · Cited by 73,173
BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325
Efficient Estimation of Word Representations in Vector Space
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013) · arXiv · doi:10.48550/arxiv.1301.3781 · Cited by 26,568
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Momentum Contrast for Unsupervised Visual Representation Learning
He, K., Fan, H., Wu, Y., Xie, S., & Girshick, R. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00975 · Cited by 13,347 · cited 1×
“…(SSL) methods to leverage unlabelled data in a similar manner as language models (Mikolov et al., 2013; Devlin et al., 2018; Radford et al., 2018, 2019) and vision models (Doersch et al., 2015; Zhang et al., 2016; Noroozi and Favaro, 2016; Wei et al., 2018; He et al., 2020; Chen et al., 2020).”
Identity mappings in deep residual networks
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_38 · Cited by 10,741 · cited 2×
“We adapted a ResNet-V2 with 18 layers and 1D convolutions (He et al., 2016) for the main trunk (feature extractor), totalling 10M parameters.”
Identity Mappings in Deep Residual Networks
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · arXiv · doi:10.48550/arxiv.1603.05027 · Cited by 10,665 · cited 2×
“We adapted a ResNet-V2 with 18 layers and 1D convolutions (He et al., 2016) for the main trunk (feature extractor), totalling 10M parameters.”
UMAP: Uniform Manifold Approximation and Projection
McInnes, L., Healy, J., Saul, N., & Großberger, L. (2018) · Journal of Open Source Software · doi:10.21105/joss.00861 · Cited by 9,826 · cited 1×
“Cluster analysis We used UMAP (McInnes et al., 2018) with default parameters for low-dimensional projections for visualization.”
UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
McInnes, L., Healy, J., & Melville, J. (2018) · arXiv · doi:10.48550/arxiv.1802.03426 · Cited by 7,101
Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
Simonyan, K., Vedaldi, A., & Zisserman, A. (2013) · arXiv · doi:10.48550/arxiv.1312.6034 · Cited by 5,370 · cited 1×
“…various LRP parameter options, LRP-0 , LRP-(cid:15) , LRP-CMP, as well as some popular off-the-shelf XAI attribution frameworks, such as saliency mapping (Simonyan et al., 2013), Guided Backpropagation (GBP) (Springenberg et al., 2014) and Integrated Gradients (IG) (Sundararajan et al., 2017).”
BackgroundBootstrap Your Own Latent: A New Approach to Self-Supervised Learning
Grill, J.-B., Strub, F., Altché, F., Tallec, C., Richemond, P. H., Buchatskaya, E., Doersch, C., Pires, B. A., Guo, Z. D., Azar, M. G., et al. (2020) · arXiv · doi:10.48550/arxiv.2006.07733 · Cited by 3,846 · cited 1×
“This prompts the use of self-supervised learning (SSL) methods to leverage unlabelled data in a similar manner as language models (Mikolov et al., 2013; Devlin et al., 2018; Radford et al., 2018, 2019) and vision models (Doersch et al., 2015; Zhang et al., 2016; Noroozi and Favaro, 2016; Wei et…”
Striving for Simplicity: The All Convolutional Net
Springenberg, J. T., Dosovitskiy, A., Brox, T., & Riedmiller, M. (2014) · arXiv · doi:10.48550/arxiv.1412.6806 · Cited by 3,447
Zhang, R., Isola, P., & Efros, A. A. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46487-9_40 · Cited by 3,178 · cited 1×
“…(SSL) methods to leverage unlabelled data in a similar manner as language models (Mikolov et al., 2013; Devlin et al., 2018; Radford et al., 2018, 2019) and vision models (Doersch et al., 2015; Zhang et al., 2016; Noroozi and Favaro, 2016; Wei et al., 2018; He et al., 2020; Chen et al., 2020).”
Unsupervised Learning of Visual Representations by Solving Jigsaw Puzzles
Noroozi, M., & Favaro, P. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46466-4_5 · Cited by 2,775 · cited 1×
“…(SSL) methods to leverage unlabelled data in a similar manner as language models (Mikolov et al., 2013; Devlin et al., 2018; Radford et al., 2018, 2019) and vision models (Doersch et al., 2015; Zhang et al., 2016; Noroozi and Favaro, 2016; Wei et al., 2018; He et al., 2020; Chen et al., 2020).”
Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition
Ordóñez, F., & Roggen, D. (2016) · Sensors · doi:10.3390/s16010115 · Cited by 2,761
On the Opportunities and Risks of Foundation Models
Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M. S., Bohg, J., Bosselut, A., Brunskill, E., et al. (2021) · arXiv · doi:10.48550/arxiv.2108.07258 · Cited by 2,676
Unsupervised Visual Representation Learning by Context Prediction
Doersch, C., Gupta, A., & Efros, A. A. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.167 · Cited by 2,629 · cited 1×
“…(SSL) methods to leverage unlabelled data in a similar manner as language models (Mikolov et al., 2013; Devlin et al., 2018; Radford et al., 2018, 2019) and vision models (Doersch et al., 2015; Zhang et al., 2016; Noroozi and Favaro, 2016; Wei et al., 2018; He et al., 2020; Chen et al., 2020).”
Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
Goyal, P., Dollár, P., Girshick, R., Noordhuis, P., Wesolowski, L., Kyrola, A., Tulloch, A., Jia, Y., & He, K. (2017) · arXiv · doi:10.48550/arxiv.1706.02677 · Cited by 2,294 · cited 1×
“To our surprise, the downstream performance did not degrade too much even when we reduced the subject data ratio from 1.0 to .25 (Figure 3b right).”
InfluentialA tutorial on human activity recognition using body-worn inertial sensors
Bulling, A., Blanke, U., & Schiele, B. (2014) · ACM Computing Surveys · doi:10.1145/2499621 · Cited by 1,668 · cited 3×
“HAR models are usually built following the Activity Recognition Chain (Bulling et al., 2014), which describes the pre-processing and segmentation of raw time series into fixed-length frames, from which features are extracted for classification.”
InfluentialDataset Shift in Machine Learning
(2008) · doi:10.7551/mitpress/9780262170055.001.0001 · Cited by 1,553 · cited 1×
“In contrast to previous works, we provide a more realistic evaluation of the utility of SSL-HAR by factoring in common issues seen in the practical use cases of pre-trained models such as domain shift and task shift (Quiñonero-Candela et al., 2008).”
Introducing a New Benchmarked Dataset for Activity Monitoring
Reiss, A., & Stricker, D. (2012) · 2012 16th International Symposium on Wearable Computers · doi:10.1109/iswc.2012.13 · Cited by 1,456
Doherty, A., Jackson, D., Hammerla, N., Plötz, T., Olivier, P., Granat, M. H., White, T., van Hees, V. T., Trenell, M. I., Owen, C. G., et al. (2017) · PLOS ONE · doi:10.1371/journal.pone.0169649 · Cited by 1,250
Self-citationHuman activity recognition with smartphone sensors using deep learning neural networks
Ronao, C. A., & Cho, S.-B. (2016) · Expert Systems with Applications · doi:10.1016/j.eswa.2016.04.032 · Cited by 1,215
Training Compute-Optimal Large Language Models
Hoffmann, J., Borgeaud, S., Mensch, A., Buchatskaya, E., Cai, T., Rutherford, E., Casas, D. D. L., Hendricks, L. A., Welbl, J., Clark, A., et al. (2022) · Advances in Neural Information Processing Systems 35 · doi:10.52202/068431-2176 · Cited by 932 · cited 1×
“The test performance for the AoT stayed at the random chance level, and the test performance for permutation dropped roughly 10 percentage points without weighted sampling.”
Training Compute-Optimal Large Language Models
Hoffmann, J., Borgeaud, S., Mensch, A., Buchatskaya, E., Cai, T., Rutherford, E., Casas, D. D. L., Hendricks, L. A., Welbl, J., Clark, A., et al. (2022) · arXiv · doi:10.48550/arxiv.2203.15556 · Cited by 929 · cited 1×
“The test performance for the AoT stayed at the random chance level, and the test performance for permutation dropped roughly 10 percentage points without weighted sampling.”
Collecting complex activity datasets in highly rich networked sensor environments
Roggen, D., Calatroni, A., Rossi, M., Holleczek, T., Forster, K., Troster, G., Lukowicz, P., Bannach, D., Pirkl, G., Ferscha, A., et al. (2010) · 2010 Seventh International Conference on Networked Sensing Systems (INSS) · doi:10.1109/inss.2010.5573462 · Cited by 788
Realistic Evaluation of Deep Semi-Supervised Learning Algorithms
Oliver, A., Odena, A., Raffel, C., Cubuk, E. D., & Goodfellow, I. J. (2018) · arXiv · doi:10.48550/arxiv.1804.09170 · Cited by 772
Layer-Wise Relevance Propagation: An Overview
Montavon, G., Binder, A., Lapuschkin, S., Samek, W., & Müller, K.-R. (2019) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-28954-6_10 · Cited by 750
Um, T. T., Pfister, F. M. J., Pichler, D., Endo, S., Lang, M., Hirche, S., Fietzek, U., & Kulić, D. (2017) · Proceedings of the 19th ACM International Conference on Multimodal Interaction · doi:10.1145/3136755.3136817 · Cited by 729 · cited 3×
“To fully appreciate the benefit of SSL in HAR, we chose the tasks that would prioritise the temporal dependencies of human motion, namely, arrow of time (AoT), permutation and time-warping (TW) (Um et al., 2017).”
Wearables and the Medical Revolution
Dunn, J., Runge, R., & Snyder, M. (2018) · Personalized Medicine · doi:10.2217/pme-2018-0044 · Cited by 654
Validation of the GENEA Accelerometer
ESLIGER, D. W., ROWLANDS, A. V., HURST, T. L., CATT, M., MURRAY, P., & ESTON, R. G. (2011) · Medicine & Science in Sports & Exercise · doi:10.1249/mss.0b013e31820513be · Cited by 625
Deep, Convolutional, and Recurrent Models for Human Activity Recognition Using Wearables
Hammerla, N. Y., Halloran, S., & Ploetz, T. (2016) · arXiv · doi:10.48550/arxiv.1604.08880 · Cited by 606
Um, T. T. (n.d.) · mediaTUM – the media and publications repository of the Technical University Munich (Technical University Munich) · Cited by 602
Activity Recognition Using a Single Accelerometer Placed at the Wrist or Ankle
MANNINI, A., INTILLE, S. S., ROSENBERGER, M., SABATINI, A. M., & HASKELL, W. (2013) · Medicine & Science in Sports & Exercise · doi:10.1249/mss.0b013e31829736d6 · Cited by 383 · cited 1×
“…for a network of the same architecture but fully trained from scratch, and a strong random forest model with tried-and-tested time series features, which has often been neglected in baseline model comparisons (Zhang et al., 2012; Mannini et al., 2013; Ellis et al., 2016; Willetts et al., 2018).”
On-body localization of wearable devices: An investigation of position-aware activity recognition
Sztyler, T., & Stuckenschmidt, H. (2016) · 2016 IEEE International Conference on Pervasive Computing and Communications (PerCom) · doi:10.1109/percom.2016.7456521 · Cited by 369
Smartphone and Smartwatch-Based Biometrics Using Activities of Daily Living
Weiss, G. M., Yoneda, K., & Hayajneh, T. (2019) · IEEE Access · doi:10.1109/access.2019.2940729 · Cited by 360
Lubitz, S. A., Faranesh, A. Z., Selvaggi, C., Atlas, S. J., McManus, D. D., Singer, D. E., Pagoto, S., McConnell, M. V., Pantelopoulos, A., & Foulkes, A. S. (2022) · Circulation · doi:10.1161/circulationaha.122.060291 · Cited by 355
Multi-task Self-Supervised Learning for Human Activity Detection
Saeed, A., Ozcelebi, T., & Lukkien, J. (2019) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3328932 · Cited by 346 · cited 4×
“Recent studies explored the utility of SSL for HAR (Saeed et al., 2019; Tang et al., 2020), but these still relied on small-scale laboratory-style datasets, hence the full potential of SSL-HAR remains unknown.”
InfluentialLearning and Using the Arrow of Time
Wei, D., Lim, J., Zisserman, A., & Freeman, W. T. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00840 · Cited by 315 · cited 1×
“…(SSL) methods to leverage unlabelled data in a similar manner as language models (Mikolov et al., 2013; Devlin et al., 2018; Radford et al., 2018, 2019) and vision models (Doersch et al., 2015; Zhang et al., 2016; Noroozi and Favaro, 2016; Wei et al., 2018; He et al., 2020; Chen et al., 2020).”
Seshadri, D. R., Davies, E. V., Harlow, E. R., Hsu, J. J., Knighton, S. C., Walker, T. A., Voos, J. E., & Drummond, C. K. (2020) · Frontiers in Digital Health · doi:10.3389/fdgth.2020.00008 · Cited by 298
Wearable devices in clinical trials: hype and hypothesis
Izmailova, E. S., Wagner, J. A., & Perakslis, E. D. (2018) · Clinical Pharmacology & Therapeutics · doi:10.1002/cpt.966 · Cited by 265
Willetts, M., Hollowell, S., Aslett, L., Holmes, C., & Doherty, A. (2018) · Scientific Reports · doi:10.1038/s41598-018-26174-1 · Cited by 250 · cited 3×
“Indeed, some studies found that earlier approaches, such as random forests, tended to generalise better than deep learning models in certain small data regimes (Willetts et al., 2018; Kwon et al., 2020).”
Self-citationA systematic review of smartphone-based human activity recognition methods for health research
Straczkiewicz, M., James, P., & Onnela, J.-P. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00514-4 · Cited by 247
Physical Activity Classification Using the GENEA Wrist-Worn Accelerometer
ZHANG, S., ROWLANDS, A. V., MURRAY, P., & HURST, T. L. (2012) · Medicine & Science in Sports & Exercise · doi:10.1249/mss.0b013e31823bf95c · Cited by 233 · cited 1×
“…for a network of the same architecture but fully trained from scratch, and a strong random forest model with tried-and-tested time series features, which has often been neglected in baseline model comparisons (Zhang et al., 2012; Mannini et al., 2013; Ellis et al., 2016; Willetts et al., 2018).”
BackgroundHip and Wrist Accelerometer Algorithms for Free-Living Behavior Classification
ELLIS, K., KERR, J., GODBOLE, S., STAUDENMAYER, J., & LANCKRIET, G. (2016) · Medicine & Science in Sports & Exercise · doi:10.1249/mss.0000000000000840 · Cited by 186 · cited 1×
“…for a network of the same architecture but fully trained from scratch, and a strong random forest model with tried-and-tested time series features, which has often been neglected in baseline model comparisons (Zhang et al., 2012; Mannini et al., 2013; Ellis et al., 2016; Willetts et al., 2018).”
BackgroundContrastive Predictive Coding for Human Activity Recognition
Haresamudram, H., Essa, I., & Plötz, T. (2021) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3463506 · Cited by 150 · cited 1×
“, 2020a), contrastive learning (Haresamudram et al., 2020b; Tang et al., 2020), and multi-task SSL (Saeed et al.”
Position-aware activity recognition with wearable devices
Sztyler, T., Stuckenschmidt, H., & Petrich, W. H. (2017) · Pervasive and Mobile Computing · doi:10.1016/j.pmcj.2017.01.008 · Cited by 148
Masked reconstruction based self-supervision for human activity recognition
Haresamudram, H., Beedu, A., Agrawal, V., Grady, P. L., Essa, I., Hoffman, J., & Plötz, T. (2020) · Proceedings of the 2020 International Symposium on Wearable Computers · doi:10.1145/3410531.3414306 · Cited by 146 · cited 2×
“The test performance for the AoT stayed at the random chance level, and the test performance for permutation dropped roughly 10 percentage points without weighted sampling.”
A Comprehensive Study of Activity Recognition Using Accelerometers
Twomey, N., Diethe, T., Fafoutis, X., Elsts, A., McConville, R., Flach, P., & Craddock, I. (2018) · Informatics · doi:10.3390/informatics5020027 · Cited by 143 · cited 1×
“Current human activity recognition (HAR) models typically rely on manual feature engineering (Twomey et al., 2018; Haresamudram et al., 2019) partly due to the very limited size of existing labelled datasets.”
A Comprehensive Study of Activity Recognition Using Accelerometers
Twomey, N., Diethe, T., Fafoutis, X., Elsts, A., McConville, R., Flach, P., & Craddock, I. (2018) · doi:10.20944/preprints201803.0147.v1 · Cited by 143
SelfHAR: Improving Human Activity Recognition through Self-training with Unlabeled Data
Tang, C. I., Perez-Pozuelo, I., Spathis, D., Brage, S., Wareham, N., & Mascolo, C. (2021) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3448112 · Cited by 140 · cited 1×
“Recent studies explored the use of SSL for HAR, such as forecasting (Taghanaki and Etemad, 2020), masked reconstruction (Haresamudram et al., 2020a), contrastive learning (Haresamudram et al., 2020b; Tang et al., 2020), and multi-task SSL (Saeed et al., 2019; Tang et al., 2021).”
Hammerla, N. Y., Kirkham, R., Andras, P., & Ploetz, T. (2013) · Proceedings of the 2013 International Symposium on Wearable Computers · doi:10.1145/2493988.2494353 · Cited by 138 · cited 1×
“Early work in HAR found some success in tree-based approaches with hand-crafted features (Hammerla et al., 2013).”
Analysis of human behavior recognition algorithms based on acceleration data
Bruno, B., Mastrogiovanni, F., Sgorbissa, A., Vernazza, T., & Zaccaria, R. (2013) · 2013 IEEE International Conference on Robotics and Automation · doi:10.1109/icra.2013.6630784 · Cited by 134
Mobile health: the power of wearables, sensors, and apps to transform clinical trials
Munos, B., Baker, P. C., Bot, B. M., Crouthamel, M., de Vries, G., Ferguson, I., Hixson, J. D., Malek, L. A., Mastrototaro, J. J., Misra, V., et al. (2016) · Annals of the New York Academy of Sciences · doi:10.1111/nyas.13117 · Cited by 118
Small, S. R., Bullock, G. S., Khalid, S., Barker, K. L., Trivella, M., & Price, A. J. (2019) · BMJ Open · doi:10.1136/bmjopen-2019-033832 · Cited by 115
The Use of Wearables in Clinical Trials During Cancer Treatment: Systematic Review
Beauchamp, U. L., Pappot, H., & Holländer-Mieritz, C. (2020) · JMIR mHealth and uHealth · doi:10.2196/22006 · Cited by 110
Assessing the State of Self-Supervised Human Activity Recognition Using Wearables
Haresamudram, H., Essa, I., & Plötz, T. (2022) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3550299 · Cited by 107 · cited 3×
“…(SSL) methods to leverage unlabelled data in a similar manner as language models (Mikolov et al., 2013; Devlin et al., 2018; Radford et al., 2018, 2019) and vision models (Doersch et al., 2015; Zhang et al., 2016; Noroozi and Favaro, 2016; Wei et al., 2018; He et al., 2020; Chen et al., 2020).”
Time--frequency analysis of biosignals
Addison, P. S., Walker, J., & Guido, R. C. (2009) · IEEE Engineering in Medicine and Biology Magazine · doi:10.1109/memb.2009.934244 · Cited by 95 · cited 1×
“…based on visualising the raw sensor signal, its analogues time-frequency representation through continuous wavelet transform (CWT) scalograms (Addison et al., 2009), as well as the time-and pretext task-localised LRP relevance scores, all with respect to observing the concurrent video…”
On the role of features in human activity recognition
Haresamudram, H., Anderson, D. V., & Plötz, T. (2019) · Proceedings of the 23rd International Symposium on Wearable Computers · doi:10.1145/3341163.3347727 · Cited by 90 · cited 1×
“Current human activity recognition (HAR) models typically rely on manual feature engineering (Twomey et al., 2018; Haresamudram et al., 2019) partly due to the very limited size of existing labelled datasets.”
Descriptive epidemiology of physical activity energy expenditure in UK adults (The Fenland study)
Lindsay, T., Westgate, K., Wijndaele, K., Hollidge, S., Kerrison, N., Forouhi, N., Griffin, S., Wareham, N., & Brage, S. (2019) · International Journal of Behavioral Nutrition and Physical Activity · doi:10.1186/s12966-019-0882-6 · Cited by 83
Daneault, J., Vergara-Diaz, G. P., Parisi, F., Admati, C., Alfonso, C., Bertoli, M., Bonizzoni, E., Carvalho, G. F., Costante, G., Fabara, E. E., et al. (2021) · Scientific Data · doi:10.1038/s41597-021-00830-0 · Cited by 73
Exploring Contrastive Learning in Human Activity Recognition for Healthcare
Tang, C. I., Perez-Pozuelo, I., Spathis, D., & Mascolo, C. (2020) · arXiv · doi:10.48550/arxiv.2011.11542 · Cited by 67
Creagh, A. P., Hamy, V., Yuan, H., Mertes, G., Tomlinson, R., Chen, W.-H., Williams, R., Llop, C., Yee, C., Duh, M. S., et al. (2022) · doi:10.1101/2022.11.18.22282305 · Cited by 64
Self-citationCreagh, A. P., Hamy, V., Yuan, H., Mertes, G., Tomlinson, R., Chen, W.-H., Williams, R., Llop, C., Yee, C., Duh, M. S., et al. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01013-y · Cited by 64 · cited 1×
“The test performance for the AoT stayed at the random chance level, and the test performance for permutation dropped roughly 10 percentage points without weighted sampling.”
Self-citationWearable accelerometer-derived physical activity and incident disease
Khurshid, S., Weng, L.-C., Nauffal, V., Pirruccello, J. P., Venn, R. A., Al-Alusi, M. A., Benjamin, E. J., Ellinor, P. T., & Lubitz, S. A. (2022) · npj Digital Medicine · doi:10.1038/s41746-022-00676-9 · Cited by 63
Wearable technology for early detection of COVID-19: A systematic scoping review
Cheong, S. H. R., Ng, Y. J. X., Lau, Y., & Lau, S. T. (2022) · Preventive Medicine · doi:10.1016/j.ypmed.2022.107170 · Cited by 59
Yuan, H., Plekhanova, T., Walmsley, R., Reynolds, A. C., Maddison, K. J., Bucan, M., Gehrman, P., Rowlands, A., Ray, D. W., Bennett, D., et al. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01065-0 · Cited by 40
Self-citationBreaking away from labels: The promise of self-supervised machine learning in intelligent health
Spathis, D., Perez-Pozuelo, I., Marques-Fernandez, L., & Mascolo, C. (2022) · Patterns · doi:10.1016/j.patter.2021.100410 · Cited by 39
Creagh, A. P., Lipsmeier, F., Lindemann, M., & Vos, M. D. (2021) · Scientific Reports · doi:10.1038/s41598-021-92776-x · Cited by 35 · cited 1×
“Contextualising layer-wise relevance propagation We applied layer-wise relevance propagation (LRP) to visually investigate the signal characteristics relevant for detecting the pretext tasks (Mon-tavon et al., 2019; Creagh et al., 2021).”
Self-citationSmall, S. R., Chan, S., Walmsley, R., Fritsch, L. V., Acquah, A., Mertes, G., Feakins, B. G., Creagh, A., Strange, A., Matthews, C. E., et al. (2023) · doi:10.1101/2023.02.20.23285750 · Cited by 22 · cited 1×
“The test performance for the AoT stayed at the random chance level, and the test performance for permutation dropped roughly 10 percentage points without weighted sampling.”
Self-citationTomasev, N., Bica, I., McWilliams, B., Buesing, L., Pascanu, R., Blundell, C., & Mitrovic, J. (2022) · arXiv · doi:10.48550/arxiv.2201.05119 · Cited by 22
Evaluating Contrastive Learning on Wearable Timeseries for Downstream Clinical OutcomesPreprint
Shah, K., Spathis, D., Tang, C. I., & Mascolo, C. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.07089 · Cited by 15
Research works printed but not linked: 3
- Chen, T., Kornblith, S., Norouzi, M. & Hinton, G. A simple framework for contrastive learning of visual representations. In International Conference on Machine Learning, 1597–1607 (PMLR, 2020).
- Radford, A. et al. Language models are unsupervised multitask learners. OpenAI Blog 1, 9 (2019).
- Haresamudram, H., Essa, I. & Plötz, T. Contrastive Predictive Coding for Human Activity Recognition. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 5, 26 (2021). · DOI · Paper page
Cited by
Creagh, A. P., Hamy, V., Yuan, H., Mertes, G., Tomlinson, R., Chen, W.-H., Williams, R., Llop, C., Yee, C., Duh, M. S., et al. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01013-y · Cited by 64
Self-citationCreagh, A. P., Hamy, V., Yuan, H., Mertes, G., Tomlinson, R., Chen, W.-H., Williams, R., Llop, C., Yee, C., Duh, M. S., et al. (2022) · doi:10.1101/2022.11.18.22282305 · Cited by 64
Self-citationChan, S., Hang, Y., Tong, C., Acquah, A., Schonfeldt, A., Gershuny, J., & Doherty, A. (2024) · Scientific Data · doi:10.1038/s41597-024-03960-3 · Cited by 61
“Our existing benchmark incorporates common methods used for HAR, future work could also benefit from leveraging more recent modeling techniques using DeepConvLSTM 40,41 , transformers 42 , and self-supervised learning 43–46 .”
Self-citationYuan, H., Plekhanova, T., Walmsley, R., Reynolds, A. C., Maddison, K. J., Bucan, M., Gehrman, P., Rowlands, A., Ray, D. W., Bennett, D., et al. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01065-0 · Cited by 40
“See Supplementary Section 1.3.1 for the full speci fi cation of the analysis.”
BackgroundInfluentialSelf-citationA Comprehensive Survey on Wearable Computing for Mental and Physical Health Monitoring
Elfouly, T., & Alouani, A. T. (2025) · Electronics · doi:10.3390/electronics14173443 · Cited by 71
“[17] [18] This bipolar disorder management using wearable technologies”.”
Metadata
- Authors
- Yuan, H., Chan, S., Creagh, A. P., Tong, C., Acquah, A., Clifton, D. A., & Doherty, A.
- Affiliations
- University of Oxford (GB)
- Corresponding author
- Doherty, Aiden
- Venue
- npj Digital Medicine
- Published
- 2024-04-12
- Volume
- 7
- Issue
- 1
- Pages
- 91-91
- ISSN
- 2398-6352 (ISSN-L)
- PMC
- PMC11015005
- OpenAlex
- W4394757986
- Cited by
- 166
- References
- 56
- Citations by year
- 2026: 78 · 2025: 57 · 2024: 26
- FWCI
- 22.43
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- creativecommons.org/licenses/by/4.0 (tdm)
- creativecommons.org/licenses/by/4.0 (vor)
- Locations
- npj Digital Medicine (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) (cc-by)
- Funders
- Wellcome Trust ([223100/Z/21/Z])
- British Heart Foundation
- Novo Nordisk
- GlaxoSmithKline
- DH | National Institute for Health Research
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
- Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Physical Activity and Health
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- human activity recognition, self-supervised learning, wearable accelerometers, UK Biobank, physical activity monitoring, unlabeled data, pre-trained models, cross-dataset generalization, sensors, limited labeled data
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 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
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
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.