Multi-task Self-Supervised Learning for Human Activity Detection
Open accessThe DOI matches its Crossref record; 85 of the 85 references it cites were found in a registry or an index.
Abstract
Deep learning methods are successfully used in applications pertaining to ubiquitous computing, pervasive intelligence, health, and well-being. Specifically, the area of human activity recognition (HAR) is primarily transformed by the convolutional and recurrent neural networks, thanks to their ability to learn semantic representations directly from raw input. However, in order to extract generalizable features massive amounts of well-curated data are required, which is a notoriously challenging task; hindered by privacy issues and annotation costs. Therefore, unsupervised representation learning (i.e., learning without manually labeling the instances) is of prime importance to leverage the vast amount of unlabeled data produced by smart devices. In this work, we propose a novel self-supervised technique for feature learning from sensory data that does not require access to any form of semantic labels, i.e., activity classes. We learn a multi-task temporal convolutional network to recognize transformations applied on an input signal. By exploiting these transformations, we demonstrate that simple auxiliary tasks of the binary classification result in a strong supervisory signal for extracting useful features for the down-stream task. We extensively evaluate the proposed approach on several publicly available datasets for smartphone-based HAR in unsupervised, semi-supervised and transfer learning settings. Our method achieves performance levels superior to or comparable with fully-supervised networks trained directly with activity labels, and it performs significantly better than unsupervised learning through autoencoders. Notably, for the semi-supervised case, the self-supervised features substantially boost the detection rate by attaining a kappa score between 0.7 - 0.8 with only 10 labeled examples per class. We get similar impressive performance even if the features are transferred from a different data source. Self-supervision drastically reduces the requirement of labeled activity data, effectively narrowing the gap between supervised and unsupervised techniques for learning meaningful representations. While this paper focuses on HAR as the application domain, the proposed approach is general and could be applied to a wide variety of problems in other areas.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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 cited source doesn’t support the claim as written.
In the citing paper
“In contrast, previous studies applied self-supervised training and fine-tuning only on the same data sources24,27,40, making it necessary to pre-train on every new dataset in practical applications.”
From Self-supervised learning for human activity recognition using 700,000 person-days of wearable data (2024) · cites this paper as [27]
Checked by a reasoning model — may be wrong.
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
“For windows of sensor data, Saeed et al. [166] apply each transformation with a probability of 50%, and passed the probabilistically transformed windows through a common convolutional encoder.”
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 Saeed et al. [166]
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
“Haresamudram et al. [81] show that the SimCLR framework [184] and multi-task self-supervised learning [166], outperform other approaches, including BYOL, SimSiam, masked reconstruction, CPC and autoencoders for HAR.”
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 [166]
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
“The established multi-task self-sup. Ref. [15] framework is ill-suited for discrete representation learning, with significant reductions in performance throughout, consistently performing worse by approx. 10% for most datasets and peaking at around 29% for Mobiact.”
From Towards Learning Discrete Representations via Self-Supervision for Wearables-Based Human Activity Recognition (2024) · cites this paper as [15]
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
“We show that our method outperforms existing self-supervised methods in HAR for UCI HAR, MotionSense, and HAPT datasets.”
From Self-supervised Human Activity Recognition by Learning to Predict Cross-Dimensional Motion (2021) · cites this paper as [19]
Experimental — accuracy not yet measured
19 more claims about this paper
Create a free account to see every claim other papers make about this one.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 1×
“These techniques have achieved indisputable empirical success across a broad spectrum of problems [4, 21, 28, 36, 43, 56, 65, 66] thanks to the increasing dataset sizes and computing power availability.”
BackgroundAdam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 3×
“We use ReLU as non-linearity in all the layers (except the output) and train a network with Adam optimizer [26] for a maximum of 30 epochs with a learning rate of 0.”
MethodInfluentialLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
(2008) · Journal of Machine Learning Research · Cited by 33,051
Pan, S. J., & Yang, Q. (2010) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2009.191 · Cited by 23,387 · cited 1×
“owledge transfer se−ings (such as features, instances, weights) from existing labeled data of some related task or domain. For a detailed review of transfer learning, we refer an interested reader to [53]. 2.4 Semi-supervised Learning Semi-supervised learning provides a compelling framework for leveraging unlabeled da…”
BackgroundInfluentialSequence to Sequence Learning with Neural Networks
Sutskever, I., Vinyals, O., & Le, Q. V. (2014) · arXiv · doi:10.48550/arxiv.1409.3215 · Cited by 17,766 · cited 1×
“These techniques have achieved indisputable empirical success across a broad spectrum of problems [4, 21, 28, 36, 43, 56, 65, 66] thanks to the increasing dataset sizes and computing power availability.”
Background(2010) · International Conference on Machine Learning · Cited by 13,995
Representation Learning: A Review and New Perspectives
Bengio, Y., Courville, A., & Vincent, P. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2013.50 · Cited by 13,001 · cited 3×
“is inherent limitation of the fully-supervised learning paradigm emphasizes the importance of unsupervised learning to leverage a large amount of unlabeled data for representation learning [8] that can be easily acquired in a real-world seing.”
BackgroundInfluentialCaruana, R. (1997) · Machine Learning · doi:10.1023/a:1007379606734 · Cited by 6,636 · cited 2×
“e goal of multi-task learning (MTL) is to enhance the learning eciency and accuracy through simultaneously optimizing multiple objectives based on shared representations and exploiting relations among the tasks [10].”
BackgroundDeepFace: Closing the Gap to Human-Level Performance in Face Verification
Taigman, Y., Yang, M., Ranzato, M., & Wolf, L. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.220 · Cited by 6,542 · cited 1×
“ese techniques have achieved indisputable empirical success across a broad spectrum of problems [4, 21, 28, 35, 43, 56, 65, 66] thanks to the increasing dataset sizes and computing power availability.”
Background
Research works printed but not linked: 7
- Baldi Pierre, 2012, Proceedings of ICML workshop on unsupervised and transfer learning. 37--49
- Chatzaki Charikleia, 2016, International Conference on Information and Communication Technologies for Ageing Well and e-Health. Springer, 100--118
- 10.5555/3104322.3104425 · doi:10.5555/3104322.3104425
- 10.5555/3305890.3305968 · doi:10.5555/3305890.3305968
- Plötz Thomas, 2011, IJCAI Proceedings-International Joint Conference on Artificial Intelligence, 22
- Razavian Narges, 2016, Machine Learning for Healthcare Conference. 73--100
- Wawrzyniak S., 2015 Federated Conference on Computer Science and Information Systems (FedCSIS). 411--416
Cited by
Items 201–300 of 345
Rey, V. F., Rezende, P. M. B., Zhou, B., Suh, S., & Lukowicz, P. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129288 · Cited by 14
Large Language Model-guided Semantic Alignment for Human Activity Recognition
Yan, H., Tan, H., Ding, Y. C., Zhou, P., Namboodiri, V., & Yang, Y. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3770652 · Cited by 31
“…methods mitigate domain shifts, they often struggle with large domain discrepancies, depend heavily on abundant source data Data augmentation [30, 35, 42, 43, 48, 52] in human activity recognition aims to increase the diversity and quality of training data, thereby enhancing the model’s…”
Adaptive Vehicle Detection in Urban Environments: A Self-learning Approach
Guerrero-Contreras, G., Balderas-Díaz, S., García-Pascual, A., & Muñoz, A. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-031-83117-1_3 · Cited by 4
OpenMAE: Efficient Masked Autoencoder for Vibration Sensing with Open-domain Data Enrichment
Hu, C., Chen, Y., Kara, D., Liu, S., Abdelzaher, T., Wu, F., & Chen, G. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3729485 · Cited by 6
Self-supervised Representation Learning for Time Series via Temporal Contrasting and Transformation
Chen, Q., Zhao, M., & Zhang, Z. (2023) · International Conference on Computer Communication and Artificial Intelligence (CCAI) · doi:10.1109/ccai57533.2023.10201298 · Cited by 4
“In the study of learning self-supervised model time series based on contrast, Saeed et al. [16]similarly designed a binary classification prior task for human activity recognition by applying several transformations to time series data and trained the model to classify between the original and…”
MethodSelf-Supervised Learning from Unlabeled IoT Data
Liu, D., & Abdelzaher, T. (2023) · Springer eBooks · doi:10.1007/978-3-031-40787-1_2 · Cited by 5
MultiHGR: Multi-Task Hand Gesture Recognition with Cross-Modal Wrist-Worn Devices
Lyu, M., Zhou, H., Guo, K., Zhou, W., Shen, X., & Gu, Y. (2024) · IEEE Conference on Computer Communications (INFOCOM) · doi:10.1109/infocom52122.2024.10621430 · Cited by 4
“Saeed et al. [23] proposed a pre-task for pretraining a model to identify which signal transformation was applied among eight options for accelerometer data.”
BackgroundSemi-supervised Human Activity Recognition with individual difference alignment
Yang, Z., Li, T., Xu, Z., Huang, Z., Cao, Y., Li, K., & Ma, J. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.126976 · Cited by 4
PRimuS: Pretraining IMU Encoders with Multimodal Self-Supervision
Das, A. M., Tang, C. I., Kawsar, F., & Malekzadeh, M. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10888874 · Cited by 19
“Thus, we also compare PRIMUS against (III) MultitaskSSL [8], a well-established self-supervised approach for IMU signals.”
Pallapothula, V. K. N., Anand, S., Bhattacharjee, S. D., & Yuan, J. (2025) · Smart Health · doi:10.1016/j.smhl.2025.100571 · Cited by 4
Zebhi, S. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-17400-8 · Cited by 4
Khaertdinov, B., & Asteriadis, S. (2023) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb57857.2023.10448965 · Cited by 5
Towards Customizable Foundation Models for Human Activity Recognition with Wearable Devices
Qiu, M., Weng, C., Fan, M., & Wu, K. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3749479 · Cited by 4
“Additionally, HAR systems incorporate vision-based [32], wireless-based [16, 21, 50, 61, 66, 67], and mobile-based [17, 24, 27, 38, 40, 41, 43, 55, 57, 59] methods.”
Self-Supervised Learning for Complex Activity Recognition Through Motif Identification Learning
Xia, Q., Morales, J., Huang, Y., Hara, T., Wu, K., Oshima, H., Fukuda, M., Namioka, Y., & Maekawa, T. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3514736 · Cited by 7
DisCovHAR: Contrastive Attention for Human Activity Recognition Under Distribution Shifts
Chen, L., Odema, M., & Faruque, M. A. A. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3551263 · Cited by 5
“D vector rotation equations for each axis [28], [43].”
InfluentialYang, Z., Li, K., Huang, Z., Xu, Z., Zhu, X., & Xiao, Y. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.112317 · Cited by 3
Wonderwall: A Virtual-to-Real Foundation Model for IMU-based HAR
Miao, S., & Chen, L. (2026) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3789688 · Cited by 9
“Our augmentation incorporates eight transformation techniques [46, 50]: Gaussian noise addition, random scaling, 3D rotation, signal inversion, time reversal, segment scrambling, temporal warping, and channel shuffling.”
Human Activity Classification Using Recurrence Plot and Residual Network
Lew, C. H., Lim, K. M., Lee, C. P., & Lim, J. Y. (2023) · IEEE Conference on Systems, Process and Control · doi:10.1109/icspc59664.2023.10420336 · Cited by 3
EarDA: Towards Accurate and Data-Efficient Earable Activity Sensing
Lyu, S., Chen, Y., Duan, D., Jia, R., & Xu, W. (2024) · doi:10.1109/cscaiot62585.2024.00005 · Cited by 15
“Thus, four datasets are chosen as they are widely utilized in previous works [14], [15].”
MethodSong, Y., Pitafi, Z. F., Dou, F., Sun, J., Zhang, X., Phillips, B. G., & Song, W. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3678514 · Cited by 5
ModelX : A Novel Transfer Learning Approach Across Heterogeneous Datasets
Dey, A., Antony, N., Dhakal, A. R., Thopalli, K., Thiagarajan, J. J., Patki, T., Marathe, A., Scogland, T., Yeom, J.-S., & Islam, T. Z. (2025) · ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC) · doi:10.1145/3731545.3731593 · Cited by 5
“…the optimal hyperparameters for the NN used in 𝑓 𝑠 , M A , and M R , we use an automated hyperparameter tuning framework named Optuna [3] with the following parameter ranges: #-hidden layers [ 3 , 100 ] , learning rate [ 1 𝑒 − 5 , 1 𝑒 − 1 ] , #-neurons [ 50 , 150 ] , with 𝑑𝑟𝑜𝑝𝑜𝑢𝑡 = 0 .”
Okadome, Y., Ata, K., Ishiguro, H., & Nakamura, Y. (2022) · Transactions of the Japanese Society for Artificial Intelligence · doi:10.1527/tjsai.37-6_b-m43 · Cited by 3
Model and Empirical Study on Multi-tasking Learning for Human Fall Detection
Nguyen, D., Pham, C., Argent, R., Caulfield, B., & Le‐Khac, N. (2024) · Vietnam Journal of Computer Science · doi:10.1142/s2196888824500040 · Cited by 5
Haresamudram, H., Tang, C. I., Suh, S., Lukowicz, P., & Plötz, T. (2024) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3675094.3677562 · Cited by 8
“A number of such methods have been designed specifically for time-series wearable sensor data, e.g., multi-task probabilistic transformation discrimination [20] and reconstruction of data [9].”
Tian, Y., Gu, Y., Lu, Q., & Kamijo, S. (2025) · Sensors · doi:10.3390/s25030649 · Cited by 8
“We redeployed Proposed method and baseline work TPN [37], MSRLSTM [10], DCL [38], ABLSTM [39], PEN [40], and PCL [41] on the Tokyo dataset to evaluate performance in the consumer behavior application scenario.”
Transformer-Based Contrastive Meta-Learning For Low-Resource Generalizable Activity Recognition
Wang, J., & Faruque, M. A. A. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889848 · Cited by 13
“Additionally, existing deep learning-based HAR require extensive labeled data to achieve reasonable performance while human-involved data are privacy-sensitive, costly to annotate, and susceptible to variations of sensors and environments [11], [12].”
Weather Recognition Using Self-supervised Deep Learning
Acuña-Escobar, D., Intriago-Pazmiño, M., & Ibarra-Fiallo, J. (2022) · Communications in computer and information science · doi:10.1007/978-3-030-99170-8_12 · Cited by 3
Applications of Self-Supervised Learning to Biomedical Signals: where are we nowPreprint
Pup, F. D., & Atzori, M. (2023) · doi:10.36227/techrxiv.22567021.v1 · Cited by 4
Self-supervised learning for early detection of neurodegenerative diseases with small data
Jiang, H. (2023) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/166402 · Cited by 9
Contrastive Self-Supervised Learning for Stress Detection from ECG Data
Rabbani, S., & Khan, N. (2023) · doi:10.32920/22734386.v1 · Cited by 3
Enhancing Traffic Flow Forecasting with Deep‐Learning and Noise Reduction Techniques
Zhao, J., Yu, L., Wang, S., & Zhang, Z. (2024) · Discrete Dynamics in Nature and Society · doi:10.1155/2024/1928189 · Cited by 3
From Vision to Motion: Translating Large-Scale Knowledge for Data-Scarce IMU Applications
Yoon, H., Cha, H., Nguyen, H. C., Gong, T., & Lee, S.-J. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3556998 · Cited by 10
Wu, X., & Zhao, B. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-9891-2_38 · Cited by 2
Frequency-Aware Masked Autoencoders for Human Activity Recognition using Accelerometers
Lorenzen, N. R., Jennum, P. J., Mignot, E., & Brink-Kjaer, A. (2025) · 2025 47th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc58623.2025.11252605 · Cited by 2
“Wearable accelerometry is a cost-effective tool for out-of-clinic health monitoring, with applications in human activity recognition (HAR) [1], [2], [3], [4], gait estimation [5], [6], and sleep monitoring [7].”
Watanabe, K., Kano, K., Hossain, T., Katayama, S., Urano, K., Yonezawa, T., & Kawaguchi, N. (2025) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3714394.3756198 · Cited by 2
“MultiTaskSSL [16] trains the model to detect augmentations, encouraging the acquisition of semantic and invariant features common to sensor data.”
Ban, L., Zhu, T., Lu, X., Qiu, Q., Han, W., Li, S., Chen, L., Wang, K. I.-K., Nie, M., & Wan, Y. (2026) · Neurocomputing · doi:10.1016/j.neucom.2026.133963 · Cited by 10
Improving crystal material property prediction with multi-view geometric graph transformer
Zhang, L., Wang, Z., (91924), X. W., (14029), X. Z., Mao, Y., Chen, K., & Wu, Y. (2026) · Nature Communications · doi:10.1038/s41467-026-73627-7 · Cited by 2
“We leverage pretraining and fi ne-tuning strategies 21,22 with MGT. Speci fi cally, a multi-task self-supervised learning 23,24 (SSL) pretraining strategy, including denoising and contrastive learning, is employed to enhance themodel ’ s capability to effectively recognize crystal structures.”
Non-Contrastive Learning-based Behavioural Biometrics for Smart IoT Devices
Jayawardena, O., Rashid, F., & Seneviratne, S. (2023) · ACM Symposium on QoS and Security for Wireless and Mobile Networks (Q2SWinet) · doi:10.1145/3616391.3622773 · Cited by 2
“We are planning to release all of our codes and data splits publicly upon the acceptance of the paper for reproducibility of our results and to stem further research in the area.”
BackgroundInfluentialYang, L., Hao, Y., & Shinoda, K. (2023) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco58844.2023.10289883 · Cited by 2
“Saeed, et al. [9] introduced the “pretrain-then-finetune” workflow into self-supervised sensor representation learning.”
BackgroundMethodInfluentialGuo, C., Zhang, Y., Chen, Y., Yang, W., Wu, Q., & Wang, Z. (2023) · IEEE Smart World Congress (SWC) · doi:10.1109/swc57546.2023.10449131 · Cited by 2
Luo, B., Qiu, Q., Zhu, T., & Liu, Z. (2024) · IFIP advances in information and communication technology · doi:10.1007/978-3-031-57919-6_7 · Cited by 2
HARFMR: Human Activity Recognition with Feature Masking and Reconstruction
Cui, W., Chen, Y., Huang, Y. Q., Liu, C., & Zhu, T. (2024) · IFIP advances in information and communication technology · doi:10.1007/978-3-031-57919-6_6 · Cited by 2
Cross-subject generalizable representation learning with class-subject dual labels for biosignals
Kim, H., Kim, J., & Kim, S. B. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.111855 · Cited by 4
Bukit, T. A., Yahya, B. N., & Lee, S. (2024) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/compsac61105.2024.00255 · Cited by 3
Weak-Annotation of HAR Datasets using Vision Foundation Models
Bock, M., Laerhoven, K. V., & Moeller, M. (2024) · ACM International Symposium on Wearable Computers (ISWC) · doi:10.1145/3675095.3676613 · Cited by 4
“Code as well as supplementary figures and results are publicly downloadable via github.com/mariusbock/weak_har.”
Ma, G., & Ye, X. (2025) · Journal of Organizational and End User Computing · doi:10.4018/joeuc.377798 · Cited by 4
“In order to overcome the problem of insufficient labeled data, SSL, as an unsupervised learning method, has been gradually applied in the field of user behavior prediction in recent years (Saeed et al., 2019; Tlijani et al., 2023).”
Unsupervised Representation Learning Method In Sensor Based Human Activity Recognition
Takenaka, K., & Hasegawa, T. (2022) · International Conference on Machine Learning and Cybernetics · doi:10.1109/icmlc56445.2022.9941334 · Cited by 2
The Context Hierarchical Contrastive Learning for Time Series in Frequency Domain
Wang, M., & Liu, J. (2023) · Communications in computer and information science · doi:10.1007/978-981-99-1639-9_4 · Cited by 2
Ghosh, S., Mandi, S., Sen, S., Mitra, B., & De, P. (2024) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642833 · Cited by 2
“In the existing literature, researchers have explored the MTL-based approaches that utilize the sensor data to reduce the annotation effort for different applications (e.g., HAR [31, 62]).”
Achieving Single-Sensor Complex Activity Recognition from Multi-Sensor Training DataPreprint
Lago, P., Matsuki, M., & Inoue, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.11284 · Cited by 5
“Our method is inspired in techniques that learn a shared, low-dimensional representation for transfer learning [2, 32, 29, 37].”
BackgroundMethodMultimodal Federated Learning.Preprint
Zhao, Y., Barnaghi, P. M., & Haddadi, H. (2021) · arXiv (Cornell University) · Cited by 4
Applications of Self-Supervised Learning to Biomedical Signals: where are we nowPreprint
Pup, F. D., & Atzori, M. (2023) · doi:10.36227/techrxiv.22567021 · Cited by 4
Applications of Self-Supervised Learning to Biomedical Signals: where are we nowPreprint
Pup, F. D., & Atzori, M. (2023) · doi:10.36227/techrxiv.22567021.v2 · Cited by 2
Oishi, N., Birch, P., Lago, P., & Roggen, D. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116367 · Cited by 1
Exploring Sensor Modalities to Capture User Behaviors for Reading Detection
Islam, M. R., Vargo, A. W., Iwata, M., Iwamura, M., & Kise, K. (2022) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2020zdl0003 · Cited by 1
Reinforcement Learning for Time Series Data with Partially Labeled Anomalies
Yun, K., & Baek, J. (2024) · International Conference on Artificial Intelligence in Information and Communication (ICAIIC) · doi:10.1109/icaiic60209.2024.10463344 · Cited by 1
Multi-Vision Services Acceleration Framework for IoT Devices
Xie, J., Ding, C., Ren, S., & Li, Y. (2024) · IEEE International Conference on Web Services (ICWS) · doi:10.1109/icws62655.2024.00107 · Cited by 1
Self-enhanced multi-task and split federated learning framework for RIS-aided cell-free systems
Urakami, T., Jia, H., Chen, N., & Okada, M. (2024) · Internet of Things · doi:10.1016/j.iot.2024.101406 · Cited by 2
Impact of Pre-training Datasets on Human Activity Recognition with Contrastive Predictive Coding
da Silva, B. E. R., Napoli, O. O., Delgado, J. V., Rocha, A. R., Boccato, L., & Borin, E. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-79035-5_21 · Cited by 3
An Analysis of Time-Frequency Consistency in Human Activity Recognition
Hecker, N., Napoli, O. O., Delgado, J., Rocha, A. R., Boccato, L., & Borin, E. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-79035-5_5 · Cited by 4
Advancing Parkinson's Disease Management Through Multi-Shared-Task Self-Supervised Signal Processing
Shuqair, M., Shahed, J. J., & Ghoraani, B. (2024) · Asilomar Conference on Signals, Systems and Computers (ACSSC) · doi:10.1109/ieeeconf60004.2024.10942713 · Cited by 1
“This study presents an innovative M-SSL methodology, extending from the previously proposed multi-task SSL [14].”
Membership Inference Against Self-supervised IMU Sensing Applications
Zhao, T., Wang, N., & Wang, X. (2025) · ACM Conference on Embedded Networked Sensor Systems (SenSys) · doi:10.1145/3715014.3722060 · Cited by 4
SelfReplay: Adapting Self-Supervised Sensory Models via Adaptive Meta-Task Replay
Yoon, H., Kwak, J., Tolera, B. A., Dai, G., Li, M., Gong, T., Lee, K., & Lee, S.-J. (2025) · ACM Conference on Embedded Networked Sensor Systems (SenSys) · doi:10.1145/3715014.3722066 · Cited by 6
“Self-supervised techniques such as predictive coding [16, 18], contrastive learning [57], and multi-task learning [52] have demonstrated effectiveness in sensory applications by reducing reliance on labeled data in pre-training.”
InfluentialGuo, K., Tang, H., Xu, T., Zhou, H., Lyu, M., Liu, Z., Wang, X., & Li, X. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3569841 · Cited by 2
Amr, O., Ahmed, A., & Atia, A. (2025) · International Conference on New Trends in Computing Sciences (ICTCS) · doi:10.1109/ictcs65341.2025.10989385 · Cited by 1
“Similar approaches have been successful in other health monitoring applications such as sleep stage scoring [12], circadian rhythm analysis [13], eating behavior detection [14], dementia-related agitation monitoring [15], and general human activity detection using self-supervised learning [16].”
Resource-efficient Gaze Estimation via Frequency-domain Multi-task Contrastive Learning
Du, L., Zhang, X., & Lan, G. (2025) · ACM Transactions on Sensor Networks · doi:10.1145/3736767 · Cited by 1
“Multi-task learning has been leveraged to learn features that are generic across multiple tasks in various research areas, such as gaze estimation [13, 65], human activity recognition [21, 41, 46], visual tracking [36], and face recognition [37].”
DML-FitAR: A Deep Metric Learning Approach for IMU-Based Fitness Activity Recognition
Li, T., Li, D., Chen, Y., Li, Z., Ma, Y., Liu, D., Feng, X., Wu, J. Y., & Guo, C. (2025) · International Conference on Multimedia Retrieval · doi:10.1145/3731715.3733311 · Cited by 1
Spatial–Temporal based Crowd Flow Prediction Using Multiscale Dynamic Residual Learning
Abideen, Z. U., Sun, X., Su, B., & Aslam, A. (2025) · Transactions in GIS · doi:10.1111/tgis.70092 · Cited by 1
“Saeed et al. (2019) presented two classifications for multi-task supervised learning based on what information should be shared.”
High-Order Moments Conditional Domain Adaptation Networks for Wearable Human Activity Recognition
Ghosh, I., Chugh, G., Faridee, A. Z. M., & Roy, N. (2025) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3746252.3761225 · Cited by 4
Federated Attention Autoencoders with a Stochastic Aggregation Scheme for Anomaly Detection
Ilić, M., Savić, M., Kurbalija, V., Ivanović, M. D., Fortino, G., & Jakovetić, D. (2025) · International Conference on Federated Learning Technologies and Applications (FLTA) · doi:10.1109/flta67013.2025.11336743 · Cited by 2
“Representational learning can be done in an unsupervised, semi-supervised, or self-supervised fashion [4], [7], [34].”
Wang, Q., & Xia, C. (2025) · International Conference on Cyberworlds (CW) · doi:10.1109/cw68232.2025.00058 · Cited by 1
“Saeed et al. [21] introduced a Transformation Prediction Network (TPN).”
Toward Wearable Sensor-Based Human Activity Recognition: A Survey
Zou, H., Chen, Z., Pan, Y., & Li, J. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3702134 · Cited by 3
“2) Self-Supervised Learning: Self-supervised learning is a type of unsupervised learning where a model is pretrained on a large amount of unlabeled data before fine-tuning it for target WHAR tasks [195], [221], [222].”
Xie, H., Xue, F., & Wang, X. (2026) · Lifetime Data Analysis · doi:10.1007/s10985-026-09721-1 · Cited by 1
“To train the encoder effectively within this multi-task framework, the model was tasked with solving three complementary temporal pretext classification problems (Yuan et al. 2024a ; Saeed et al. 2019).”
Pellatt, L., & Roggen, D. (2022) · Frontiers in Computer Science · doi:10.3389/fcomp.2022.914330 · Cited by 2
“TPN (Saeed et al., 2019) IMU, altitude HHAR (Stisen et al.”
BackgroundTowards a New Multi-tasking Learning Approach for Human Fall Detection
Nguyen, D., Pham, C., Argent, R., Caulfield, B., & Le‐Khac, N. (2023) · Lecture notes in networks and systems · doi:10.1007/978-3-031-36886-8_5 · Cited by 2
Advanced System for Enhancing Location Identification through Human Pose and Object Detection
Kevin, M. A., Crespo, J., Gómez, J., & Álfaro, C. (2023) · Machines · doi:10.3390/machines11080843 · Cited by 1
“A novel approach to self-supervised sensor representation learning for smartphone-based human activity recognition is presented in [12].”
BackgroundHaresamudram, H., Tang, C. I., Suh, S., Lukowicz, P., & Plötz, T. (2023) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3594739.3605102 · Cited by 3
“This family of approaches, which are based on ‘Self-Supervised Learning’, has heralded a paradigm change within the community, with a gold rush of new methods aiming to make economic use of the publicly available unlabeled datasets for learning effective representations [2, 4–6, 10, 11].”
MethodUse of Domain Labels during Pre-Training for Domain-Independent WiFi-CSI Gesture Recognition
van Berlo, B., Verhoeven, R., & Meratnia, N. (2023) · Sensors · doi:10.3390/s23229233 · Cited by 3
Lim, S., Lee, Y. J., & Kim, S. B. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.111464 · Cited by 2
Zlatintsi, A., Filntisis, P. P., Efthymiou, N., Garoufis, C., Retsinas, G., Sounapoglou, T., Maglogiannis, I. G., Tsanakas, P. D., Smyrnis, N., & Maragos, P. A. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10433929 · Cited by 1
“This abundance of sensory data has actually kickstarted the development of several applications focused on health monitoring, as well as other tasks, such as emotional well-being, sleep tracking, and physical activity detection [7, 8].”
BackgroundPreliminary Investigation of SSL for Complex Work Activity Recognition in Industrial Domain via MoIL
Xia, Q., Maekawa, T., Morales, J., Hara, T., Oshima, H., Fukuda, M., & Namioka, Y. (2024) · IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) · doi:10.1109/percomworkshops59983.2024.10503195 · Cited by 1
“Multi-task self-supervision [13] .”
Domain Adaptation in Human Activity Recognition through Self-Training
Kfari, M. K. A., & Lüdtke, S. (2024) · ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp) · doi:10.1145/3675094.3678465 · Cited by 3
Xia, K., Li, W., Shao, Y., & Lu, S. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681664 · Cited by 3
“• Self-supervised Learning (SSL) [32].”
Shuqair, M., Shahed, J. J., & Ghoraani, B. (2024) · Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) · doi:10.1109/embc53108.2024.10782431 · Cited by 1
Multi-Positive Sample Quantum Contrastive Learning for Human Activity Recognition
Ren, Y., Wang, D., An, L., Mao, S., & Wang, X. (2024) · Globecom · doi:10.1109/globecom52923.2024.10901296 · Cited by 2
“Inspired multi-task self-supervised learning [11], we believe that using only two types of augmentation is insufficient to cover the full range of real-world situations in contrastive learning.”
Saga: Capturing Multi-granularity Semantics from Massive Unlabelled IMU Data
Li, Y., Hu, F., Zhu, H., Zhang, S., Zhang, L., Chang, S., & Guo, M. (2025) · IEEE International Conference on Distributed Computing Systems (ICDCS) · doi:10.1109/icdcs63083.2025.00090 · Cited by 1
“TPN [13] first tries to pre-train on IMU data by introducing multiple transformations on IMU data to use these transformations as supervision.”
Gangopadhyay, A., Morris, M., Saboury, B., Siegel, E. L., & Yesha, Y. (2020) · Digital Government Research and Practice · doi:10.1145/3428092 · Cited by 1
“Self-supervised learning in the form of rotation augmentation has shown to improve the model generalizability in a number of computer vision [3, 4, 8, 11] and human activity recognition [2, 9] tasks.”
BackgroundImproving Pain Management in Patients with Sickle Cell Disease Using Machine Learning Techniques
Yang, F. (2020) · CORE Scholar (Wright State University) · Cited by 1
Quality Assessment of Crowdwork via Eye Gaze: Towards Adaptive Personalized Crowdsourcing
Islam, M. R., Shun, N., Vargo, A. W., Iwata, M., Matsubara, M., Morishima, A., & Kise, K. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-85616-8_8 · Cited by 1
Wearable-based Classification of Running Styles with Deep LearningPreprint
Taghanaki, S. R., Rainbow, M. J., & Etemad, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.00594 · Cited by 2
“More recently, self-supervised solutions have been proposed in [16], [17], [18], and [19] to reduce reliance on labels and learn more generalized representations by means of pretext training with a number of transformations applied to wearable data.”
Consistency Training of Multi-exit Architectures for Sensor DataPreprint
Saeed, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.13192 · Cited by 1
Self-citationStudy on Temperature (τ) Variation for SimCLR based Activity RecognitionPreprint
Kumar, P. (2021) · Preprints.org · doi:10.20944/preprints202107.0138.v2
“TPN [35] is incorporated as a base encoder for the HAR systems to suit the needs of a comparatively lightweight neural network architecture.”
MethodTowards robust and label-efficient time series representation learning
Emadeldeen, A. I. A. E. (2023) · doi:10.32657/10356/170673 · Cited by 1
Building generalizable deep learning solutions for mobile sensing
Xu, H. (2024) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/174809 · Cited by 1
Temporal Distance Based Self-Supervised Learning for Human Activity Recognition
Bigelli, L., Contoli, C., Kania, N., & Lattanzi, E. (2025) · 2025 Sixth International Conference on Intelligent Data Science Technologies and Applications (IDSTA) · doi:10.1109/idsta66210.2025.11202839
“Saeed et al. [11] proposed a general approach to self-supervised learning by applying various transformations to the signal, namely noise, scaling, rotation, negation, horizontal flipping, permutation time-warping and channel shuffling , and training a network with a common feature extractor…”
Haresamudram, H., Tang, C. I., Suh, S., Lukowicz, P., & Ploetz, T. (2024) · arXiv · Cited by 31
“…to contrastive approaches in computer vision tasks, Haresamudram et al. [77] showed that the SimCLR framework [176] and multi-task self-supervised learning [158], outperformed other approaches, including BYOL, SimSiam, masked reconstruction, Contrastive Predictive Coding (CPC) and autoencoders.”
InfluentialInvestigating Enhancements to Contrastive Predictive Coding for Human Activity Recognition
Haresamudram, H., Essa, I., & Ploetz, T. (2022) · arXiv · Cited by 27
“Multi-task self supervision [19] first introduced selfsupervised learning to human activity recognition from sensor data.”
BackgroundMethodInfluentialCalatrava-Nicolás, F. M., Stoyanov, T., & Martinez Mozos, O. (2026) · Scientific Reports · doi:10.1038/s41598-026-64846-5
“Single-task transformation prediction 25 : We train the encoder with a single proxy task that performs binary classification of whether an input window has been transformed by a rotation or not.”
InfluentialSelf-supervised Deep Learning for Reading Activity ClassificationPreprint
Islam, M. R., Sakamoto, S., Yamada, Y., Vargo, A. W., Iwata, M., Iwamura, M., & Kise, K. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.03598
“[58] employs simple signal transformations such as ipping to produce the pretext task for sensor data.”
BackgroundMethodInfluentialPersonalization Models for Human Activity Recognition with Distribution Matching-Based Metrics
Nguyen, H. T., Kwon, H., Haresamudram, H., Peterson, A. F., & Plötz, T. (2021) · Communications in computer and information science · doi:10.1007/978-981-16-0575-8_4
Metadata
- Authors
- Saeed, A., Ozcelebi, T., & Lukkien, J.
- Affiliations
- Eindhoven University of Technology (NL)
- Eindhoven University of Technology, Eindhoven, The Netherlands
- ORCID
- Saeed, Aaqib
- Venue
- Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
- Published
- 2019-06-21
- Volume
- 3
- Issue
- 2
- Pages
- 1-30
- ISSN
- 2474-9567 (ISSN-L)
- OpenAlex
- W3099025572
- Cited by
- 346
- References
- 85
- Citations by year
- 2026: 20 · 2025: 65 · 2024: 82 · 2023: 71 · 2022: 49 · 2021: 42 · 2020: 17
- FWCI
- 11.94
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies (publishedVersion, public-domain)
- TU/e Research Portal (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- TU/e Research Portal (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Funders
- Horizon 2020 Framework Programme (737422)
- 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, self-supervised learning, unsupervised feature learning, multi-task learning, temporal convolutional network, smartphone sensing, semi-supervised learning, transfer learning, unlabeled 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.
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.