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
Time-Series Representation Learning via Temporal and Contextual Contrasting
Eldele, E., Ragab, M., Chen, Z., Wu, M., Kwoh, C. K., Li, X., & Guan, C. (2021) · Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2021/324 · Cited by 696
Self-Supervised ECG Representation Learning for Emotion Recognition
Sarkar, P., & Etemad, A. (2022) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2020.3014842 · Cited by 383
“A similar approach was used in [19] for human activity recognition.”
BackgroundMethodInfluentialDeep Gait Recognition: A Survey
Sepas-Moghaddam, A., & Etemad, A. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3151865 · Cited by 307
“eral loss functions [203], can provide the network with more representative features [204], [205], [206].”
BackgroundSelf-Supervised Contrastive Representation Learning for Semi-Supervised Time-Series Classification
Eldele, E., Ragab, M., Chen, Z., Wu, M., Kwoh, C.-K., Li, X., & Guan, C. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3308189 · Cited by 224
“The first approach is to apply generic regularizations, e.g., ensembling and reconstruction.”
Self-supervised learning for human activity recognition using 700,000 person-days of wearable data
Yuan, H., Chan, S., Creagh, A. P., Tong, C., Acquah, A., Clifton, D. A., & Doherty, A. (2024) · npj Digital Medicine · doi:10.1038/s41746-024-01062-3 · Cited by 166
“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.”
InfluentialContrastive 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
“Previous works such as [31, 63, 69] have demonstrated how unlabeled data can be utilized to learn useful representations for wide ranging tasks, including identifying kitchen activities [11], activity tracking in car manufacturing [71], classifying every day activities such as walking or running [4, 10, 51, 66], and m…”
BackgroundMethodInfluentialMasked 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
“Results for [30] comparable to original publication, yet not identical due to original implementation not being released, and details being omitted in [30].”
BackgroundMethodInfluentialSelf-Supervised Learning for ECG-Based Emotion Recognition
Sarkar, P., & Etemad, A. (2020) · ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp40776.2020.9053985 · Cited by 146
“Self-supervised learning is becoming popular in the field of computer vision [9, 12], natural language processing [13], speech and signal processing [14, 15], and others.”
BackgroundMethodInfluentialXu, H., Zhou, P., Tan, R., Li, M., & Shen, G. (2021) · Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems · doi:10.1145/3485730.3485937 · Cited by 150
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
“In human activity recognition, due to the difficulty in collecting labeled sensor data, semi-supervised learning has been actively researched [3, 10, 39].”
BackgroundMethodInfluentialTime Series Change Point Detection with Self-Supervised Contrastive Predictive Coding
Deldari, S., Smith, D. V., Xue, H., & Salim, F. D. (2021) · Proceedings of the Web Conference 2021 · doi:10.1145/3442381.3449903 · Cited by 136
“In recent years, self-supervised representation learning has been used to capture informative and compact representations of video [1, 33], image [8, 21], text[35], and time series [17, 31, 39, 40] data.”
MethodAdaptive Memory Networks With Self-Supervised Learning for Unsupervised Anomaly Detection
Zhang, Y., Wang, J., Chen, Y., Yu, H., & Qin, T. (2022) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3139916 · Cited by 165
“Specifi-cally, we propose to utilize six signal transformations inspired by [45], which are described as follows: 1) Noise: As noisy sensor signal may exist in the real world, adding noise to signal can help model learn more robust features against the noise.”
MethodMultimodal Federated Learning on IoT Data
Zhao, Y., Barnaghi, P. M., & Haddadi, H. (2022) · IEEE/ACM International Conference on Internet-of-Things Design and Implementation (IoTDI) · doi:10.1109/iotdi54339.2022.00011 · Cited by 170
ColloSSL: Collaborative Self-Supervised Learning for Human Activity Recognition
Jain, Y., Tang, C. I., Min, C., Kawsar, F., & Mathur, A. (2022) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3517246 · Cited by 113
“This design choice is inspired by prior work on self-supervised learning in HAR [46] and allows us to do a fair comparison of our”
BackgroundMethodResultInfluentialSelf-Supervised EEG Emotion Recognition Models Based on CNN
Wang, X., Ma, Y., Cammon, J., Fang, F. Z., Gao, Y., & Zhang, Y. (2023) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2023.3263570 · Cited by 131
“In the literature [28], the author used multi-task convolutional neural network for self-supervised learning, learned the features of pre-processed data through multi-task learning, and studied human activity detection.”
BackgroundMethodWu, Y., Daoudi, M., & Amad, A. (2024) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2023.3263907 · Cited by 106
“The followings are brief descriptions of these models: Supervised learning-based methods: (cid:2) SimpDCNN [27]: it is a simple convolutional network consisting of three convolutional blocks with kernel sizes of 24, 16 and 8, each followed by a ReLU activation and a dropout layer.”
MethodResultInfluentialFederated Self-Supervised Learning of Multisensor Representations for Embedded Intelligence
Saeed, A., Salim, F. D., Ozcelebi, T., & Lukkien, J. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3009358 · Cited by 106
“The prominent approaches for learning from traditional input modalities include colorization of grayscale images [34], predicting relative location of an image patch [25], audio–visual synchronization [2], temporal alignment in videos through cycle-consistency [35], word2vec (and other variants) [3], signal transforma…”
BackgroundSelf-citationAssessing 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
“3 and 4, as it has been shown to be very effective compared to grid searches [3], while also not consuming extensive computational resources.”
BackgroundMethodInfluentialCOCOA: Cross Modality Contrastive Learning for Sensor Data
Deldari, S., Xue, H., Saeed, A., Smith, D. V., & Salim, F. D. (2022) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3550316 · Cited by 92
“For UCIHAR dataset, as suggested by [63] and [73], we used windows with 400 samples.”
BackgroundMethodSelf-citationFederated Self-Supervised Learning of Multisensor Representations for Embedded Intelligence
Saeed, A., Salim, F. D., Özçelebi, T., & Lukkien, J. J. (2024) · Figshare · Cited by 92
Self-citationWearable Sensor-Based Human Activity Recognition with Hybrid Deep Learning Model
Luwe, Y. J., Lee, C. P., & Lim, K. M. (2022) · Informatics · doi:10.3390/informatics9030056 · Cited by 92
“It can be summarized that the accuracy of deep learning techniques is sorted from low to high as 79.86%, 82.50%, 83.30%, 85.59%, 85.75%, 89.00% and 89.08% of BERT [28], CAE [29], multi-task self-supervision CNN [30], DeepConvLSTM [31], CNN [32], DCNN [33] and CPC [34], in ascending order.”
BackgroundMethodInfluentialContrastive Self-supervised Learning for Sensor-based Human Activity Recognition
Khaertdinov, B., Ghaleb, E., & Asteriadis, S. (2021) · 2021 IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb52358.2021.9484410 · Cited by 95
“For this purpose, similarly to [7, 8, 17], we conduct a series of experiments when k ∈ { 1 , 2 , 5 , 10 , 25 , 50 , 100 } labeled examples per class are randomly sampled from the training set and used to train the supervised and SSL models and compare their performance.”
BackgroundMethodInfluentialWang, Q., Luo, H., Xiong, H., Men, A., Zhao, F., Xia, M. Y., & Ou, C. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3016146 · Cited by 84
“Saeed et al. [19] proposed multitask self-supervised learning to recognize human activities.”
BackgroundA survey on unsupervised learning for wearable sensor-based activity recognition
Ige, A. O., & Noor, M. H. M. (2022) · Applied Soft Computing · doi:10.1016/j.asoc.2022.109363 · Cited by 99
Towards federated unsupervised representation learning
van Berlo, B., Saeed, A., & Ozcelebi, T. (2020) · Proceedings of the Third ACM International Workshop on Edge Systems, Analytics and Networking · doi:10.1145/3378679.3394530 · Cited by 79
“A starting point for these parameters was retrieved from [21].”
BackgroundMethodInfluentialSelf-citationSensor Data Augmentation by Resampling in Contrastive Learning for Human Activity Recognition
Wang, J., Zhu, T., Gan, J., Chen, L. L., Ning, H., & Wan, Y. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3214198 · Cited by 76
Zlatintsi, A., Filntisis, P. P., Garoufis, C., Efthymiou, N., Maragos, P. A., Menychtas, A., Maglogiannis, I. G., Tsanakas, P. D., Sounapoglou, T., Kalisperakis, E., et al. (2022) · Sensors · doi:10.3390/s22197544 · Cited by 85
“, emotional well-being [5–8], sleep tracking [9,10], eating [11], agitation [12] and physical activity detection [13,14].”
BackgroundWhat Makes Good Contrastive Learning on Small-Scale Wearable-based Tasks?
Qian, H., Tian, T., & Miao, C. (2022) · Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining · doi:10.1145/3534678.3539134 · Cited by 67
“According to the wearable sensors’ characteristics, various pretext tasks can be constructed, such as masked reconstruction [15], motion forecasting [37], and data transformation predictions [31].”
BackgroundHong, Z., Li, Z., Zhong, S., Lyu, W., Wang, H., Ding, Y., He, T., & Zhang, D. (2024) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3659597 · Cited by 67
“We utilize UCI [56], Shoaib [63], Motion [46], and HHAR [65], to evaluate different methods, as they are widely used in previous HAR studies [7, 50, 58, 82, 83, 85].”
BackgroundMethodIntegration of federated learning with IoT for smart cities applications, challenges, and solutions
Ghadi, Y. Y., Mazhar, T., Shah, S. F. A., Haq, I., Ahmad, W., Ouahada, K., & Hamam, H. (2023) · PeerJ Computer Science · doi:10.7717/peerj-cs.1657 · Cited by 79
SigRep: Toward Robust Wearable Emotion Recognition With Contrastive Representation Learning
Dissanayake, V., Seneviratne, S., Rana, R. K., Wen, E., Kaluarachchi, T. I., & Nanayakkara, S. C. (2022) · IEEE Access · doi:10.1109/access.2022.3149509 · Cited by 70
Human Activity Recognition Using Wearable Sensors: Review, Challenges, Evaluation Benchmark
Abdel-Salam, R., Mostafa, R., & Hadhood, M. (2021) · Communications in Computer and Information Science · doi:10.1007/978-981-16-0575-8_1 · Cited by 60
Chan, 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 .”
IMUGPT 2.0: Language-Based Cross Modality Transfer for Sensor-Based Human Activity Recognition
Leng, Z., Bhattacharjee, A., Rajasekhar, H., Zhang, L., Bruda, E., Kwon, H., & Plötz, T. (2024) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3678545 · Cited by 61
“The research community has developed a range of approaches to specifically tackle the sparse data problem, including self-supervised learning [26–28, 67, 73], few-shot learning [21], semi-supervised learning [10], prototypical learning [9, 17], adversarial learning [8, 41], and transfer learning…”
Practically Adopting Human Activity Recognition
Xu, H., Zhou, P., Tan, R., & Li, M. (2023) · Proceedings of the 29th Annual International Conference on Mobile Computing and Networking · doi:10.1145/3570361.3613299 · Cited by 55
“■ DCNN [63] designs a CNN-based HAR model that out-performs ■ TPN [40] learns features from unlabeled data by recognizing ■ LIMU-GRU [62] learns representations by a self-supervised ■ HDCNN [17] handles domain shift by minimizing the ■ FM [4] minimizes the feature distance across domains by ■ XHAR…”
MethodNegative Selection by Clustering for Contrastive Learning in Human Activity Recognition
Wang, J., Zhu, T., Chen, L. L., Ning, H., & Wan, Y. (2023) · IEEE Internet of Things Journal · doi:10.1109/jiot.2023.3239945 · Cited by 52
“Some self-supervised learning works are applied to human activity recognition based on sensor data, such as Multitask SSL [33], CAE [38], Masked Reconstruction [39], and CPCHAR [40].”
MethodInfluentialTransfer Learning Approach for Human Activity Recognition Based on Continuous Wavelet Transform
Pavliuk, O., Mishchuk, M., & Strauß, C. (2023) · Algorithms · doi:10.3390/a16020077 · Cited by 69
Self-supervised Learning for Electroencephalogram: A Systematic Survey
Weng, W., Gu, Y., Guo, S., Ma, Y., Yang, Z., Liu, Y., & Chen, Y. (2025) · ACM Computing Surveys · doi:10.1145/3736574 · Cited by 98
“Subsequently, by minimizing the binary cross-entropy pretext task loss, the model learns the distinctive spatial features of different channels and retains essential information beneficial for various downstream tasks.”
Liu, D., Wang, T., Liu, S., Wang, R., Yao, S., & Abdelzaher, T. (2021) · 2021 International Conference on Computer Communications and Networks (ICCCN) · doi:10.1109/icccn52240.2021.9522151 · Cited by 44
Unsupervised Human Activity Representation Learning with Multi-task Deep Clustering
Ma, H., Zhang, Z., Li, W., & Lu, S. (2021) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3448074 · Cited by 48
Chen, H., Lundberg, S. M., Erion, G., Kim, J. H., & Lee, S.-I. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00536-y · Cited by 48
“Other broad categories of approaches involve data augmentations of accelerometer data aimed towards improving generalization [60, 61] and contrastive learning that focuses on similarity of negative and positive pairs of samples [62 – 65].”
ESPRESSO: Entropy and ShaPe awaRe timE-Series SegmentatiOn for Processing Heterogeneous Sensor Data
Deldari, S., Smith, D. V., Sadri, A., & Salim, F. (2020) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3411832 · Cited by 48
“Consequently, unsupervised and 100:2 • Shohreh Deldari, Daniel V. Smith, Amin Sadri, Flora D. Salim self-supervised techniques have gained a lot of attention [39].”
Cross-Domain HAR: Few-Shot Transfer Learning for Human Activity Recognition
Thukral, M., Haresamudram, H., & Plötz, T. (2024) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3704921 · Cited by 70
“Self-supervised learning is a form of unsupervised learning, which derives supervisory signals for downstream performance by creating auxiliary tasks [19, 24, 40] on unlabeled data.”
InfluentialRecent Advances in Pedestrian Navigation Activity Recognition: A Review
Wang, Q., Luo, H., Wang, Q., Sun, L., Ma, Z., Zhang, C., Fu, M., & Zhao, F. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3153610 · Cited by 50
Cross-Domain HAR: Few Shot Transfer Learning for Human Activity Recognition
Thukral, M., Haresamudram, H., & Ploetz, T. (2023) · arXiv · doi:10.48550/arxiv.2310.14390 · Cited by 45
“Self-supervised learning is a form of unsupervised learning, which derives supervisory signals for downstream performance by creating auxiliary tasks [19, 24, 40] on unlabeled data.”
InfluentialApplications of Self-Supervised Learning to Biomedical Signals: A Survey
Pup, F. D., & Atzori, M. (2023) · IEEE Access · doi:10.1109/access.2023.3344531 · Cited by 43
Task Programming: Learning Data Efficient Behavior Representations
Sun, J. J., Kennedy, A., Zhan, E., Anderson, D. J., Yue, Y., & Perona, P. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00290 · Cited by 40
Xing, Y., Lv, C., Cao, D., & Velenis, E. (2021) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2021.103288 · Cited by 50
Self-Supervised Learning for Label- Efficient Sleep Stage Classification: A Comprehensive Evaluation
Eldele, E., Ragab, M., Chen, Z., Wu, M., Kwoh, C. K., & Li, X. (2023) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2023.3245285 · Cited by 60
“In the context of time-series applications, a new classification auxiliary task was defined in [28], [29] by generating several views to the signals using augmentations such as: adding noise, rotation, and scaling .”
Label-Efficient Time Series Representation Learning: A Review
Eldele, E., Ragab, M., Chen, Z., Wu, M., Kwoh, C. K., & Li, X. (2024) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2024.3430236 · Cited by 69
Transfer Learning in Sensor-Based Human Activity Recognition: A Survey
Dhekane, S. G., & Ploetz, T. (2025) · ACM Computing Surveys · doi:10.1145/3717608 · Cited by 42
“Conventional: Activelearning[132],Co-training[64],EnsembleLearning[62,165,166,233],HMM[206] NeuralNetworkBased: (a)Architectures: LSTM[240],AE[162],GraphAEs[137],Bi-GAN[187],DBN[163] (b)LossFunctions: (c)TrainingStrategies: Adversariallearning[187] Conventional: Self-training[202],Co-training[130],Activelearning[6,39,…”
InfluentialContrastive Self-Supervised Learning for Stress Detection from ECG Data
Rabbani, S., & Khan, N. (2022) · Bioengineering · doi:10.3390/bioengineering9080374 · Cited by 52
“Furthermore, these are fundamental augmentations that are widely used in literature for time series representation learning [22].”
BackgroundIMU2Doppler: Cross-Modal Domain Adaptation for Doppler-based Activity Recognition Using IMU Data
Bhalla, S., Goel, M., & Khurana, R. (2021) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3494994 · Cited by 41
“…Activity Recognition (HAR), previous works have proposed a range of techniques including self-supervision based unsupervised transfer learning [40], few shot learning [19], substructure-level matching based domain adaptation [34], adaptive spatial-temporal transfer learning (ASTTL) [37] and…”
BackgroundSelf-supervised transfer learning of physiological representations from free-living wearable data
Spathis, D., Perez-Pozuelo, I., Brage, S., Wareham, N. J., & Mascolo, C. (2021) · Proceedings of the Conference on Health, Inference, and Learning · doi:10.1145/3450439.3451863 · Cited by 40
Semi-supervised Federated Learning for Activity RecognitionPreprint
Zhao, Y., Liu, H., Li, H., Barnaghi, P. M., & Haddadi, H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.00851 · Cited by 76
“[27] show that, in HAR, training autoencoders with unlabelled time-series data to extract useful representations can help improve the performance of down-stream tasks including supervised learning.”
BackgroundSense and Learn: Self-supervision for omnipresent sensors
Saeed, A., Ungureanu, V., & Gfeller, B. (2021) · Machine Learning with Applications · doi:10.1016/j.mlwa.2021.100152 · Cited by 34
Self-citationMachine Learning in Tourism: A Brief Overview
Egger, R. (2022) · Tourism on the verge · doi:10.1007/978-3-030-88389-8_6 · Cited by 49
Enhancing Human Activity Recognition in Smart Homes with Self-Supervised Learning and Self-Attention
Chen, H., Gouin-Vallerand, C., Bouchard, K., Gaboury, S., Couture, M., Bier, N., & Giroux, S. (2024) · Sensors · doi:10.3390/s24030884 · Cited by 50
“The augmentation module applied five selected augmentation techniques mentioned in [40,54] to the ambient sensor data.”
MethodApplying Self-Supervised Representation Learning for Emotion Recognition Using Physiological Signals
Quispe, K. G. M., Utyiama, D. M. D. S., Santos, E. M. D., Oliveira, H. A. B. F., & Souto, E. (2022) · Sensors · doi:10.3390/s22239102 · Cited by 51
“These signal transformations have already been used in human activity recognition problems [94] and for emotions [5].”
BackgroundTransfer learning and its extensive appositeness in human activity recognition: A survey
Ray, A., & Kolekar, M. H. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.122538 · Cited by 39
An active semi-supervised deep learning model for human activity recognition
Bi, H., Perelló-Nieto, M., Santos‐Rodríguez, R., Flach, P. A., & Craddock, I. J. (2022) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-022-03768-2 · Cited by 39
“Following (Saeed et al. 2019; Haresamudram et al. 2020), the data from participant 106 is used for testing, the data from participant 105 for validation and the rest for training.”
BackgroundMethodInfluentialContrastive Accelerometer–Gyroscope Embedding Model for Human Activity Recognition
Koo, I., Park, Y., Jeong, M., & Kim, C. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3222825 · Cited by 40
“The tasks include signal reconstruction via an autoencoder [33], masked signal reconstruction [34], and prediction of applied transformation [35].”
MethodSelf-supervised Contrastive Learning of Multi-view Facial Expressions
Roy, S., & Etemad, A. (2021) · INTERNATIONAL CONFERENCE ON MULTIMODAL INTERACTION · doi:10.1145/3462244.3479955 · Cited by 46
Spathis, D., & Kawsar, F. (2024) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocae090 · Cited by 102
Zheng, W., Yan, L., Gou, C., & Wang, F. (2021) · Information Fusion · doi:10.1016/j.inffus.2021.10.009 · Cited by 38
WePos: Weak-supervised Indoor Positioning with Unlabeled WiFi for On-demand Delivery
Guo, B., Zuo, W., Wang, S., Lyu, W., Hong, Z., Ding, Y. C., He, T., & Zhang, D. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3534574 · Cited by 63
“Saeed et al. [33] 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 and learn a multi-task temporal convolutional network to recognize transformations applied on an input signal.”
BackgroundLearning on the Rings: Self-Supervised 3D Finger Motion Tracking Using Wearable Sensors
Zhou, H., Lu, T., Liu, Y., Zhang, S., & Gowda, M. (2022) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3534587 · Cited by 42
“Unlike vision where there is a huge number of high-quality training datasets for developing robust ML algorithms, the wearable devices based applications lack such large-scale training data because of challenges in annotation and labeling [38, 58, 104].”
BackgroundBuilding towards Automated Cyberbullying Detection: A Comparative Analysis
Al-Harigy, L. M., Al-Nuaim, H. A., Moradpoor, N., & Tan, Z. (2022) · Computational Intelligence and Neuroscience · doi:10.1155/2022/4794227 · Cited by 32
“(iii) It narrows the gap between supervised and unsupervised learning techniques for learning meaningful representations by reducing the requirement of labelled activity data, effectively [97].”
BackgroundKhaertdinov, B., Asteriadis, S., & Ghaleb, E. (2022) · IEEE Transactions on Biometrics Behavior and Identity Science · doi:10.1109/tbiom.2022.3180591 · Cited by 40
“Another type of SSL approaches previously used for the problem is transformation networks which aim to recognise a set of transformations applied to each time-window in a batch using a multi-head output layer [8].”
BackgroundMethodHaresamudram, H., Tang, C. I., Suh, S., Lukowicz, P., & Plötz, T. (2025) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3729467 · Cited by 33
“…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.”
InfluentialHuman Activity Recognition Using Wearable Sensors Based on Image Classification
Zebhi, S. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3174280 · Cited by 36
Context-aware mutual learning for semi-supervised human activity recognition using wearable sensors
Qu, Y., Tang, Y., Yang, X., Wen, Y., & Zhang, W. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.119679 · Cited by 35
Cheng, D., Zhang, L., Qin, L., Wang, S., Wu, H., & Song, A. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3373019 · Cited by 31
Kwon, H., Abowd, G. D., & Plötz, T. (2021) · Sensors · doi:10.3390/s21248337 · Cited by 30
“Saeed et al. [41] proposed a multi-task pretext task to predict eight different types of data transformations applied to the input data.”
MethodMa, S., Chen, J., & Ho, J. W. K. (2023) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2023.107906 · Cited by 35
Logacjov, A., & Bach, K. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.107478 · Cited by 41
Saha, U., Saha, S., Kabir, M. T., Fattah, S. A., & Saquib, M. (2024) · IEEE Sensors Letters · doi:10.1109/lsens.2024.3423340 · Cited by 36
Qin, X., Wang, J., Ma, S., Lu, W., Zhu, Y., Xie, X., & Chen, Y. (2023) · Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining · doi:10.1145/3580305.3599360 · Cited by 29
“[38] verified these transformations are feasible and practical for HAR.”
BackgroundMethodThe Empathetic Car: Exploring Emotion Inference via Driver Behaviour and Traffic Context
Liu, S., Koch, K., Zhou, Z., Föll, S., He, X., Menke, T., Fleisch, E., & Wortmann, F. (2021) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3478078 · Cited by 36
“We leveraged multi-task learning approach, a method that has been proven to be useful in recognising emotions or activities [63, 68], to build a neural network to predict all emotions at the same time.”
MethodZhou, B., Geißler, D., Faulhaber, M., Gleiß, C., Zahn, E. F., Ray, L. S. S., Gamarra, D. F., Rey, V. F., Suh, S., Bian, S., et al. (2023) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3580883 · Cited by 50
Multi-task self-supervised time-series representation learning
Choi, H., & Kang, P. (2024) · Information Sciences · doi:10.1016/j.ins.2024.120654 · Cited by 34
Diffusion-Model-Based Contrastive Learning for Human Activity Recognition
Xiao, C., Han, Y., Yang, W. P., Hou, Y., Shi, F., & Chetty, K. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3429245 · Cited by 34
“Saeed et al. [66] proposed a multitask self-supervised approach that employs a multitask temporal convolutional network to learn generalizable information from the sensing data.”
BackgroundSemi-Supervised Contrastive Learning for Human Activity Recognition
Liu, D., & Abdelzaher, T. (2021) · International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT) · doi:10.1109/dcoss52077.2021.00019 · Cited by 32
MyDJ: Sensing Food Intakes with an Attachable on Your Eyeglass Frame
Shin, J., Lee, S.-J., Gong, T., Yoon, H., Roh, H., Bianchi, A., & Lee, S.-J. (2022) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3491102.3502041 · Cited by 34
“While recent sensor-based applications use complex neural layers (e.g., CNN, autoencoders) for artificial feature engineering [59, 70], we use extracted features to minimize the power and memory consumption.”
MethodTaghanaki, S. R., Rainbow, M. J., & Etemad, A. (2023) · IEEE Transactions on Human-Machine Systems · doi:10.1109/thms.2023.3303438 · Cited by 31
“Selfsupervised learning [10], [13], [14] is one of these techniques which has been proposed to tackle the problem by exploiting auxiliary tasks for pre-training, also referred to as pretext tasks, relying only on the unlabeled data.”
BackgroundMethodInfluentialSelf-Supervised Time Series Representation Learning by Inter-Intra Relational ReasoningPreprint
Fan, H., Zhang, F., & Gao, Y. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.13548 · Cited by 48
“…Saeed et al., 2020), or multi-task learning based methods that predict different hand-crafted features such as MFCCs, prosody, and waveform (Pascual et al., 2019a; Ravanelli et al., 2020), and different transformations of raw signal (Sarkar & Etemad, 2020; Saeed et al., 2019) have emerged recently.”
Investigating Enhancements to Contrastive Predictive Coding for Human Activity Recognition
Haresamudram, H., Essa, I., & Plötz, T. (2023) · 2023 IEEE International Conference on Pervasive Computing and Communications (PerCom) · doi:10.1109/percom56429.2023.10099197 · Cited by 27
“Multi-task self supervision [19] first introduced selfsupervised learning to human activity recognition from sensor data.”
BackgroundMethodInfluentialA benchmark for computational analysis of animal behavior, using animal-borne tags
Hoffman, B., Cusimano, M., Baglione, V., Canestrari, D., Chevallier, D., DeSantis, D. L., Jeantet, L., Ladds, M., Maekawa, T., Mata‐Silva, V., et al. (2024) · Movement Ecology · doi:10.1186/s40462-024-00511-8 · Cited by 37
Gullapalli, B. T., Carreiro, S., Chapman, B. P., Ganesan, D., Sjoquist, J., & Rahman, T. (2021) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3478107 · Cited by 37
“However, with recent machine learning advancements, research has shifted toward using self-supervised [74, 75] and unsupervised [48, 97] based methods to optimize the use of unlabeled data.”
BackgroundSelf-supervised Learning for Accelerometer-based Human Activity Recognition: A Survey
Logacjov, A. (2024) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3699767 · Cited by 40
Spatial-Temporal Masked Autoencoder for Multi-Device Wearable Human Activity Recognition
Miao, S., Chen, L., & Hu, R. (2023) · Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies · doi:10.1145/3631415 · Cited by 23
“Hand-crafted auxiliary tasks design additional objectives based on heuristics to provide the model with supervisory signals [61, 63, 76].”
BackgroundMethodFederated Meta-Learning with Attention for Diversity-Aware Human Activity Recognition
Shen, Q., Feng, H., Song, R., Song, D., & Xu, H. (2023) · Sensors · doi:10.3390/s23031083 · Cited by 31
“[49] propose a self-supervised technique for feature learning from sensory data.”
CroSSL: Cross-modal Self-Supervised Learning for Time-series through Latent Masking
Deldari, S., Spathis, D., Malekzadeh, M., Kawsar, F., Salim, F. D., & Mathur, A. (2024) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3616855.3635795 · Cited by 56
“…annotating, and managing extensive labeled datasets, self-supervised learning (SSL), which learns from unlabeled data, has been investigated [27]: by de � ning an arti � cial task, known as a pretext task, where the supervisory signal is automatically generated from unlabelled data,…”
MethodInfluentialInvariant Feature Learning for Sensor-based Human Activity Recognition
Hao, Y., Zheng, R., & Wang, B. (2021) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2021.3064252 · Cited by 29
“handle simple and complex activities with different task-specific layers on top of invariant features across them; whereas self-supervised learning is utilized through training an invariant feature extractor that is capable of extracting features behind various signal distortions in [32].”
BackgroundSelfPAB: large-scale pre-training on accelerometer data for human activity recognition
Logacjov, A., Herland, S., Ustad, A., & Bach, K. (2024) · Applied Intelligence · doi:10.1007/s10489-024-05322-3 · Cited by 35
“One of the first attempts to apply SSL to improve accelerometer-based HAR was made by Saeed et al. [35].”
InfluentialChen, J., Hu, Y., Garg, L., Gadekallu, T. R., Srivastava, G., & Wang, W. W. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3430297 · Cited by 30
Artificial neural networks for human activity recognition using sensor based dataset
Geravesh, S., & Rupapara, V. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13716-z · Cited by 24
“Aresearch work [26] is motivated by the success of joint learning, to solve multiple self-supervised tasks.”
BackgroundSemi-Supervised Learning for Wearable-based Momentary Stress Detection in the Wild
Yu, H., & Sano, A. (2023) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3596246 · Cited by 68
“…unlabeled samples with contrastive constraints and consistency regularization have also been examined on wearable sensor data for human activities recognition (HAR) tasks [2, 21, 28, 36, 44, 45] and emotion classification [38, 49] learned representations to supervised learning with labeled data.”
BackgroundSelf-supervised Human Activity Recognition by Learning to Predict Cross-Dimensional Motion
Rahimi Taghanaki, S., Rainbow, M. J., & Etemad, A. (2021) · 2021 International Symposium on Wearable Computers · doi:10.1145/3460421.3480417 · Cited by 22
“For IMU-based HAR, [19] and [8] have used self-supervised learning.”
BackgroundMethodInfluentialTemporal Feature Alignment in Contrastive Self-Supervised Learning for Human Activity Recognition
Khaertdinov, B., & Asteriadis, S. (2022) · IEEE International Joint Conference on Biometrics (IJCB) · doi:10.1109/ijcb54206.2022.10007984 · Cited by 28
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.