Time-Series Representation Learning via Temporal and Contextual Contrasting
Open accessThe DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index; 9 of the 21 sources its claims cite were checked for support.
Abstract
Learning decent representations from unlabeled time-series data with temporal dynamics is a very challenging task. In this paper, we propose an unsupervised Time-Series representation learning framework via Temporal and Contextual Contrasting (TS-TCC), to learn time-series representation from unlabeled data. First, the raw time-series data are transformed into two different yet correlated views by using weak and strong augmentations. Second, we propose a novel temporal contrasting module to learn robust temporal representations by designing a tough cross-view prediction task. Last, to further learn discriminative representations, we propose a contextual contrasting module built upon the contexts from the temporal contrasting module. It attempts to maximize the similarity among different contexts of the same sample while minimizing similarity among contexts of different samples. Experiments have been carried out on three real-world time-series datasets. The results manifest that training a linear classifier on top of the features learned by our proposed TS-TCC performs comparably with the supervised training. Additionally, our proposed TS-TCC shows high efficiency in few-labeled data and transfer learning scenarios. The code is publicly available at https://github.com/emadeldeen24/TS-TCC.
Overview
CoChat Summary
TL;DR
TS-TCC is an unsupervised time-series representation learning framework that combines weak/strong augmentations, a cross-view temporal contrasting module, and a contextual contrasting module. It learns representations that perform comparably to supervised training with a linear classifier and are effective in few-label and transfer settings.
Key findings
- On linear evaluation, TS-TCC achieved HAR ACC 90.37 ± 0.34 / MF1 90.38 ± 0.39, Sleep-EDF ACC 83.00 ± 0.71 / MF1 73.57 ± 0.74, and Epilepsy ACC 97.23 ± 0.10 / MF1 95.54 ± 0.08, outperforming SSL-ECG, CPC, and SimCLR and matching or exceeding supervised training on two of three datasets.
- In semi-supervised settings, TS-TCC fine-tuning reached around 70% and 90% for HAR and Epilepsy with only 1% labeled data, and with 10% labeled data achieved performance comparable to supervised training with 100% labeled data on the three datasets.
- In fault-diagnosis transfer learning, TS-TCC fine-tuning averaged 67.82% accuracy versus 63.83% for supervised pretraining, outperforming supervised in 8 of 12 cross-domain scenarios and improving by about 4% overall.
- Ablation showed the cross-view prediction task improved performance by more than 5% on HAR and about 1% on Sleep-EDF and Epilepsy; adding contextual contrasting further improved results.
Methods
- The framework creates two views per sample using weak augmentation (jitter-and-scale) and strong augmentation (permutation-and-jitter), encodes them with a 3-block convolutional encoder, and applies a Transformer-based autoregressive temporal contrasting module with a cross-view prediction task plus a contextual contrasting loss.
- Experiments used UCI HAR, Sleep-EDF, Epilepsy, and a fault diagnosis dataset; data were split 60%/20%/20% for train/validation/test, with subject-wise split for Sleep-EDF, and experiments were repeated 5 times with 5 seeds.
- Evaluation included linear classifier on frozen features, semi-supervised training with 1%, 5%, 10%, 50%, and 75% labeled data, cross-domain transfer learning, ablation, and sensitivity analysis; training used 40 epochs, Adam with lr 3e-4, batch size 128 (32 for few-label), and dataset-specific augmentation/Transformer hyperparameters.
Limitations
- Not stated in the paper: the paper does not include an explicit limitations section.
- Evident from the text: augmentation hyperparameters must be chosen carefully according to the nature of the time-series data, and the main evaluation covers three time-series datasets plus one fault-diagnosis transfer dataset.
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 · 2 Not valid · 5 Not checked · 43 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, the pretext tasks can limit the generality of the learned representations.”
What the sources say
- Not checked — no full text of the sourcecites Noroozi and Favaro, 2016 source
- Partly supportedcites Gidaris et al., 2018 sourceExperimental — accuracy not yet measured
The Introduction says representations learned through earlier self-supervised pretext tasks had not matched supervised representations in performance, which supports a qualified concern about their effectiveness. It also calls them good alternatives for transfer to other vision tasks; the paper does not establish that the pretext tasks themselves limit the representations’ generality. No page number is available in our copy. (LLM, conf=0.72)
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.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“First, they may not be able to address the temporal dependencies of data, which are key characteristics of time-series [Franceschi et al. , 2019].”
What the sources say
- Partly supportedcites Franceschi et al., 2019 sourceExperimental — accuracy not yet measured
Section 3, p. 3, uses time-based sampling and an unsupervised time-based criterion for learning time-series representations, supporting the importance of temporal information. But neither that section nor the discussion of other domains in Related Work (p. 2) says that image-based contrastive methods may fail to address temporal dependencies. (LLM, conf=0.76)
hedged- 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.
- Partly supported
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“However, these image-based contrastive learning methods are not able to work well on time-series data for the following reasons.”
What the sources say
- Could not verifycites Franceschi et al., 2019 sourceExperimental — accuracy not yet measured
Related Work (p. 2) says triplet losses have seen little use for time series and identifies limitations of certain existing time-series methods. Unsupervised Training (p. 3) explains why the authors adapt a triplet loss to time series, but neither passage assesses whether image-based contrastive learning methods work well on time-series data. Searches for image-based contrastive methods and their suitability for time series found no passage addressing that claim. (LLM, conf=0.70)
plain- 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.
- Could not verify
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, the above two methods are proposed for specific applications and they are not generalizable to other time-series data.”
What the sources say
- Not checked — no full text of the sourcecites Mohsenvand et al., 2020
- Partly supportedcites Cheng et al., 2020 sourceExperimental — accuracy not yet measured
Our copy's Related Work (Sec. 2, p. 2) does characterize prior self-supervised/contrastive work as tied to particular modalities or applications — pretext tasks shown effective for EEG [8] and ECG [9], and Hyvärinen & Morioka's time-contrastive learning [14] 'demonstrated for magnetoencephalography (MEG)' — and the Introduction (Sec. 1, pp. 1-2) adds that applying contrastive learning to biosignals 'requires careful design of appropriate data transformations and consideration of intersubject variability' and that it is 'unclear whether the same techniques used to learn representations for language and vision domains are effective for biosignals'. That supports only the 'proposed for specific applications' half. The categorical part of the claim — that existing EEG/ECG contrastive methods 'are not generalizable to other time-series data' — is nowhere asserted; the paper instead hedges about transfer in the opposite direction (vision/language → biosignals) and states in its Conclusion (Sec. 5, p. 8) that its own subject-aware contrastive approach 'can be applied to other biosignals, such as signals from the eye (EOG) or muscles (EMG)'. So the citing sentence overstates the source by turning a hedged, domain-specificity motivation into a blanket non-generalizability verdict. (LLM, conf=0.68)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Additionally, our proposed TS-TCC shows high efficiency in few-labeled data and transfer learning scenarios.”
plain- 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.
- Rests on an assumption not shown here.
32 more claims
Create a free account to see all 58 claims and every source verdict.
- Probably not a substantial claim. 12
- It may state two separate claims as one. 12
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 says what the claim attributes to it.
In the citing paper
“In the time domain, TS-TCC [116] and its extended version CA-TCC [117] designed two time series data augmentation techniques, one is strong augmentation (permutation-and-jitter), and the other is weak augmentation (jitter-and-scale).”
From Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects (2024) · cites this paper as [116]
The cited source says what the claim attributes to it.
In the citing paper
“TS-TCC [116] and its extended version CA-TCC [117] designed a cross prediction task, which uses the context of x T 1 to predict the target in x T 2 , and vice versa uses the context of x T 2 to predict the target in x T 1 .”
From Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects (2024) · cites this paper as [116]
The cited source says what the claim attributes to it.
In the citing paper
“However, CoST and BTSF do not modify the frequency representation, while TF-C [125] directly augments the time series data through frequency perturbations, which has achieved better performance than TS2Vec [118] and TS-TCC [116].”
From Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects (2024) · cites this paper as [116]
The cited source says what the claim attributes to it.
In the citing paper
“Additionally, most SSCL-based classification models emphasize that data augmentation also has a direct impact on the final classification accuracy [116], [118], [122].”
From Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects (2024) · cites this paper as [116]
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
§ 3.2 Temporal Contrasting · p. 3
“The contrastive loss tries to minimize the dot product between the predicted representation and the true one of the same sample, while maximizing the dot product with the other samples N t,k within the minibatch.”
§ 3.1 Time-Series Data Augmentation · p. 3
“Contrastive methods try to maximize the similarity among different views of the same sample, while minimizing its similarity with other samples.”
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
Exploiting Generative AI to Scale up Intelligent Tutoring Systems
Jakubův, J., Chvalovský, K., Goertzel, Z., Kaliszyk, C., Olšák, M., Piotrowski, B., Schulz, S., Suda, M., & Urban, J. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.itp.2023.19 · Cited by 76,658
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
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
A simple framework for contrastive learning of visual representations
Chen, T., Kornblith, S., Norouzi, M., & Hinton, G. (2020) · arXiv · doi:10.48550/arxiv.2002.05709 · Cited by 10,857
Representation Learning with Contrastive Predictive Coding
Oord, A. V. D., Li, Y., & Vinyals, O. (2018) · arXiv · doi:10.48550/arxiv.1807.03748 · Cited by 6,242
Bootstrap 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
Exploring Simple Siamese Representation Learning
Chen, X., & He, K. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.01549 · Cited by 3,725
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
Research works printed but not linked: 1
- [Aggarwal et al. , 2019] Karan Aggarwal, Shafiq R. Joty, Luis Fern´ andez-Luque, and Jaideep Srivastava. Adversarial unsupervised representation learning for activity timeseries. AAAI , 2019.
Cited by
Diffusion Models: A Comprehensive Survey of Methods and Applications
Yang, L., Zhang, Z., Song, Y., Hong, S., Xu, R., Zhao, Y., Zhang, W., Cui, B., & Yang, M.-H. (2024) · ACM Computing Surveys · doi:10.1145/3626235 · Cited by 1,704
“Time series data are widely used with many important real-world applications [54, 152, 228, 243].”
BackgroundTS2Vec: Towards Universal Representation of Time Series
Yue, Z., Wang, Y., Duan, J., Yang, T., Huang, C., Tong, Y., & Xu, B. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i8.20881 · Cited by 850
“TS-TCC (Eldele et al. 2021) encourages consistency of different data augmentations.”
BackgroundMethodInfluentialTSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecasting
Ekambaram, V., Jati, A., Nguyen, N., Sinthong, P., & Kalagnanam, J. (2023) · Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining · doi:10.1145/3580305.3599533 · Cited by 278
Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects
Zhang, K., Wen, Q., Zhang, C., Cai, R., Jin, M., Liu, Y., Zhang, J. Y., Liang, Y., Pang, G., Song, D., et al. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3387317 · Cited by 295
“TS-TCC [116] and its extended version CA-TCC [117] designed a cross prediction task, which uses the context of x T 1 to predict the target in x T 2 , and vice versa uses the context of x T 2 to predict the target in x T 1 .”
MethodInfluentialDeep Learning for Time Series Classification and Extrinsic Regression: A Current Survey
Mohammadi Foumani, N., Miller, L., Tan, C. W., Webb, G. I., Forestier, G., & Salehi, M. (2024) · ACM Computing Surveys · doi:10.1145/3649448 · Cited by 268
“The model’s accuracy was also improved by pre-training the parameters on massive unlabeled data.”
BackgroundKim, J., Kim, H., Kim, H., Lee, D., & Yoon, S. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11223-9 · Cited by 263
Self-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
“This journal paper is an extended version of our previous work [11].”
Self-citationDeep learning for time series forecasting: a survey
Kong, X., Chen, Z., Liu, W., Ning, K., Zhang, L., Muhammad Marier, S., Liu, Y., Chen, Y., & Xia, F. (2025) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-025-02560-w · Cited by 211
Self-citationDeep Contrastive Representation Learning With Self-Distillation
Xiao, Z., Xing, H., Zhao, B., Qu, R., Luo, S., Dai, P., Li, K., & Zhu, Z. (2023) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2023.3304948 · Cited by 161
“Like most existing time series classification algorithms [29], [30], [31], [72], DCRLS-based classification uses a two-step method.”
BackgroundMethodInfluentialLLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters
Chang, C., Wang, W.-Y., Peng, W.-C., & Chen, T.-F. (2023) · arXiv · doi:10.48550/arxiv.2308.08469 · Cited by 146
Fully-Connected Spatial-Temporal Graph for Multivariate Time-Series Data
Wang, Y., Xu, Y., Yang, J., Wu, M., Li, X., Xie, L., & Chen, Z. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i14.29500 · Cited by 134
Self-citationLLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters
Chang, C., Wang, W., Peng, W.-C., & Chen, T.-F. (2025) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3719207 · Cited by 452
“TS-TCC [8] produces two different views via strong and weak augmentations and enhances representations through contrastive learning, focusing on the temporal and contextual differences between these views.”
TimeCLR: A self-supervised contrastive learning framework for univariate time series representation
Yang, X., Zhang, Z., & Cui, R. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.108606 · Cited by 114
Contrastive autoencoder for anomaly detection in multivariate time series
Hao, Z., Yu, K., Zhang, X., Wu, G., & Yazidi, A. (2022) · Information Sciences · doi:10.1016/j.ins.2022.07.179 · Cited by 122
Time Series Contrastive Learning with Information-Aware Augmentations
Luo, D., Cheng, W. D., Wang, Y., Xu, D., Ni, J., Yu, W., Zhang, X., Liu, Y., Chen, Y., Chen, H., et al. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i4.25575 · Cited by 174
COCOA: 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
“customized and extended the similar CL framework on the ubiquitous sensor and time-series data[18, 22, 63, 70, 73, 78].”
BackgroundMethodInfluentialA Survey on Time-Series Pre-Trained Models
Ma, Q., Liu, Z., Zheng, Z., Huang, Z., Zhu, S., Yu, Z., & Kwok, J. T.-Y. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3475809 · Cited by 182
“Eldele et al. [110] used the autoregressive model’s predictedvaluesaspositivesamplepairsforcontrastivelearning, thusenablingthemodeltocapturetemporaldependenciesofthe timeseries.”
BackgroundMethodInfluentialSelf-Supervised Autoregressive Domain Adaptation for Time Series Data
Ragab, M., Eldele, E., Chen, Z., Wu, M., Kwoh, C. K., & Li, X. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3183252 · Cited by 111
Self-citationSelf-Supervised Time Series Representation Learning via Cross Reconstruction Transformer
Zhang, W., Yang, L. Q., Geng, S., & Hong, S. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3292066 · Cited by 151
ADATIME: A Benchmarking Suite for Domain Adaptation on Time Series Data
Ragab, M., Eldele, E., Tan, W. L., Foo, C.-S., Chen, Z., Wu, M., Kwoh, C. K., & Li, X. (2023) · CRC Press eBooks · doi:10.1145/3587937 · Cited by 123
“• 1-dimensional convolutional neural network (1D-CNN): consists of three convolutional blocks, where each block consists of a 1D-Convolutional network, a BatchNorm layer, a non-linearity ReLU activation function, and finally, a MaxPooling layer [4, 9].”
Self-citationAnsari, M. Y., Yaqoob, M., Ishaq, M., Flushing, E. F., Mangalote, I. A. C., Dakua, S. P., Aboumarzouk, O. M., Righetti, R., & Qaraqe, M. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11259-x · Cited by 87
“Eldele et al. (2021) introduce temporal contextual contrastive learning for enhanced representation of time series signals.”
Deep Contrastive One-Class Time Series Anomaly Detection
Wang, R., Liu, C., Mou, X., Gao, K., Guo, X., Pin, L., Wo, T., & Liu, X. (2023) · Society for Industrial and Applied Mathematics eBooks · doi:10.1137/1.9781611977653.ch78 · Cited by 83
Contrastive Domain Adaptation for Time-Series Via Temporal Mixup
Eldele, E., Ragab, M., Chen, Z., Wu, M., Kwoh, C. K., & Li, X. (2023) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2023.3293473 · Cited by 80
“2) Feature Encoder: We adopted a convolutional neural network as our feature encoder [11], [49].”
BackgroundMethodInfluentialSelf-citationADAST: Attentive Cross-Domain EEG-Based Sleep Staging Framework With Iterative Self-Training
Eldele, E., Ragab, M., Chen, Z., Wu, M., Kwoh, C. K., Li, X., & Guan, C. (2022) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2022.3189695 · Cited by 74
“We followed the design of the feature extractor proposed in [35] which consists of three blocks, such that each block consists of a 1D-convolution layer, a batch normalization layer, a non-linear ReLU activation and a MaxPooling layer, as shown in Fig.”
MethodSelf-citationWang, Y., Wu, M., Li, X., Xie, L., & Chen, Z. (2023) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2023.3241896 · Cited by 60
“The historical information and the relations between sources should be analyzed for various downstream tasks, such as prediction [1], [2], e.g., machine remaining useful life (RUL) prediction, and classification [3], [4], e.g., human activity recognition (HAR) and sleep stage classification (SSC).”
BackgroundMethodSelf-citationUnsupervised seizure identification on EEG
Potter, İ. Y., Garner, R., Lai, M., & Duncan, D. (2021) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2021.106604 · Cited by 57
MHCCL: Masked Hierarchical Cluster-Wise Contrastive Learning for Multivariate Time Series
Meng, Q., Qian, H., Liu, Y., Cui, L., Xu, Y., & Shen, Z. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i8.26098 · Cited by 52
Intra-Inter Subject Self-Supervised Learning for Multivariate Cardiac Signals
Xiang, L., Ng, D., Hong, S., & Feng, M. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i4.20376 · Cited by 73
“(Mohsenvand, Izadi, and Maes 2020; Eldele et al. 2021) proposed self-supervised learning frameworks that tested with brain waves data.”
BackgroundSVP-T: A Shape-Level Variable-Position Transformer for Multivariate Time Series Classification
Zuo, R., Li, G., Choi, B. K. K., Bhowmick, S. S., Mah, D. N., & Wong, G. L. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i9.26359 · Cited by 55
“Recently, there have been some successes in time series applications, such as forecasting (Wu et al. 2020; Lim et al. 2021) and representation learning (Eldele et al. 2021).”
BackgroundMethodMICOS: Mixed supervised contrastive learning for multivariate time series classification
Hao, S., Wang, Z., Afanasiev, A. D., Yuan, J., & Zhang, W. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.110158 · Cited by 55
Explainable time series anomaly detection using masked latent generative modeling
Lee, D.-S., Malacarne, S., & Aune, E. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.110826 · Cited by 72
ECG-FM: an open electrocardiogram foundation model
McKeen, K., Masood, S., Toma, A., Rubin, B., & Wang, B. (2025) · JAMIA Open · doi:10.1093/jamiaopen/ooaf122 · Cited by 186
Adversarial Spatiotemporal Contrastive Learning for Electrocardiogram Signals
Wang, N., Feng, P., Ge, Z., Zhou, Y., Zhou, B., & Wang, Z. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3272153 · Cited by 82
“Choosing an appropriate augmentation method is essential for the success of contrastive learning [13].”
BackgroundMethodInfluentialSelf-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
“Also, TS-TCC proposed temporal and contextual contrasting approaches to learn instance-wise representations about the sleep EEG data [10].”
InfluentialSelf-citationLabel-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
Self-citationSelf-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
“By contrasting negative and positive pairs, the model can capture the age-related brain features to improve the generalizability of generated representations [134].”
Yang, L., Zhang, X., Zhu, F., Wang, Z., & Zhang, X. (2026) · Journal of Industrial Information Integration · doi:10.1016/j.jii.2026.101076 · Cited by 44
PatchAD: A Lightweight Patch-Based MLP-Mixer for Time Series Anomaly Detection
Zhong, Z., Yu, Z., Yang, Y., Wang, W. W., Yang, K., & Chen, C. L. P. (2025) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2025.3596745 · Cited by 89
“Recently, studies such as [12] and [43] have explored integrating contrastive learning into time series representation learning.”
InfluentialTemporal-Frequency Co-training for Time Series Semi-supervised Learning
Liu, Z., Ma, Q., Ma, P., & Wang, L. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i7.26072 · Cited by 57
“Further, we find that LPDeepSSL, TS-TCC, and MTL perform worse than Supervised, which may be attributable to the large bias introduced by their designed strategies on the learning of unlabeled time series data.”
BackgroundMethodInfluentialTimeMAE: Self-Supervised Representations of Time Series with Decoupled Masked Autoencoders
Cheng, M., Tao, X., Liu, Z., Liu, Q., Zhang, H., Zhang, R., & Chen, E. (2026) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3773966.3778007 · Cited by 128
“In contrast, less computation consumption is required in discriminative methods [25], whose main solution is pulling positive examples together while pushing negative examples away [40].”
InfluentialA Stochastic Recurrent Encoder Decoder Network for Multistep Probabilistic Wind Power Predictions
Zheng, Z., & Zhang, Z. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3234130 · Cited by 39
“It has been successfully applied to benefit various downstream supervised learning tasks in different application domains including image classification [55], speech recognition [53], as well as time series classi-fication and regression [56].”
MethodAnalysis of Augmentations for Contrastive ECG Representation Learning
Soltanieh, S., Etemad, A., & Hashemi, J. (2022) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn55064.2022.9892600 · Cited by 52
“A model called TS-TCC was proposed in [53] which learned time-series data representations by”
MethodWang, S. (2026) · Energy and Buildings · doi:10.1016/j.enbuild.2026.117233 · Cited by 38
Sun, H., Huang, Y., Han, L., Fu, C., Liu, H., & Long, X. (2023) · Computers & Security · doi:10.1016/j.cose.2023.103570 · Cited by 44
Li, H., Cao, P., Wang, X., Li, Y., Yi, B., & Huang, M. (2024) · Advanced Engineering Informatics · doi:10.1016/j.aei.2024.102517 · Cited by 40
TS-Fastformer: Fast Transformer for Time-series Forecasting
Lee, S., Hong, J., Liu, L., & Choi, W. (2023) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3630637 · Cited by 52
Multivariate time series classification with crucial timestamps guidance
Zhang, D., Gao, J., & Li, X. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124591 · Cited by 37
Self-citationSeries2vec: similarity-based self-supervised representation learning for time series classification
Foumani, N. M., Tan, C. W., Webb, G. I., Rezatofighi, H., & Salehi, M. (2024) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-024-01043-w · Cited by 53
“The contrasting contextual module is then built upon the contexts from the temporal contrasting module and aims to maximize similarity among contexts of the same sample while minimizing similarity among contexts of different samples (Eldele et al. 2021).”
InfluentialSimpleTS: An Efficient and Universal Model Selection Framework for Time Series Forecasting
Yao, Y., Li, D., Jie, H., Jie, H., Li, T., Chen, J., Wang, J., Li, F., & Gao, Y. (2023) · Proceedings of the VLDB Endowment · doi:10.14778/3611540.3611561 · Cited by 43
“DeepAR num_cells = [20, 30, 40, 50, 60] num_layers = [1, 2, 3, 4, 5] LSTM num_cells = [20, 30, 40, 50, 60] num_layers = [1, 2, 3, 4, 5] Transformer inner_ff_dim_scale = [2, 4, 6, 8, 10] num_heads = [4, 6, 8, 10, 12] Prophet changepoint_prior_scale = [0.”
BackgroundInfluentialTransferable Self-Supervised Instance Learning for Sleep Recognition
Zhao, A., Wang, Y., & Li, J. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3176751 · Cited by 30
“We also compare three SOTA unsupervised methods for sleep recognition published in 2021 [59]–[61], the results on dataset e1 are very stable, slightly inferior to our proposed method.”
BackgroundMethodResultInfluentialLin, Y., Wang, Y., Zhang, M., Wang, Z., Zhang, H., & Zhao, M. (2024) · Advanced Engineering Informatics · doi:10.1016/j.aei.2024.102471 · Cited by 33
An adversarial contrastive autoencoder for robust multivariate time series anomaly detection
Yu, J., Gao, X., Zhai, F., Li, B., Xue, B., Fu, S., Chen, L., & Meng, Z. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.123010 · Cited by 36
Senane, Z., Cao, L., Buchner, V., Tashiro, Y., You, L., Herman, P. A., Nordahl, M. G., Tu, R., & von Ehrenheim, V. (2024) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3637528.3671673 · Cited by 58
Feng, Y., Zhang, X., & Yang, C. (2024) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2024.3447108 · Cited by 32
“Nowadays, most time series models only focus on the time-domain modeling, ignoring the frequency domain information [27].”
BackgroundFeng, H., Shen, Q., Song, R., Shi, L., & Xu, H. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.121296 · Cited by 32
Sleep Stage Classification Via Multi-View Based Self-Supervised Contrastive Learning of EEG
Zhao, C., Wu, W., Zhang, H., Zhang, R., Zheng, X., & Kong, X. (2024) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2024.3432633 · Cited by 33
A survey on self-supervised methods for visual representation learning
Uelwer, T., Robine, J., Wagner, S. S., Höftmann, M., Upschulte, E., Konietzny, S. G., Behrendt, M., & Harmeling, S. (2025) · Machine Learning · doi:10.1007/s10994-024-06708-7 · Cited by 79
A contrastive learning-based framework for wind power forecast
Zhu, N., Dai, Z., Wang, Y., & Zhang, K. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.120619 · Cited by 32
ETP: Learning Transferable ECG Representations via ECG-Text Pre-Training
Liu, C., Wan, Z., Cheng, S. L., Zhang, M., & Arcucci, R. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10446742 · Cited by 43
“Consequently, our ECG invariant learning goal is defined as: In Eq 6, the temperature hyper-parameter σ 2 retains its value of 0.07 when considering the overall loss objective L SSL .”
BackgroundMethodInfluentialFEAT: A general framework for feature-aware multivariate time-series representation learning
Kim, S.-W., Chung, E., & Kang, P. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.110790 · Cited by 30
Robust multimodal federated learning for incomplete modalities
Yu, S., Wang, J., Hussein, W., & Hung, P. C. K. (2023) · Computer Communications · doi:10.1016/j.comcom.2023.12.003 · Cited by 34
Li, Y., Xie, S., Wang, J., Zhang, J., & Yan, H. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3318686 · Cited by 26
Zhang, P., Hu, W., Cao, W., Chen, L., & Wu, M. (2024) · Neural Computing and Applications · doi:10.1007/s00521-024-09548-7 · Cited by 25
Self-citationMulGad: Multi-granularity contrastive learning for multivariate time series anomaly detection
Xiao, B.-W., Xing, H.-J., & Li, C.-G. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103008 · Cited by 26
Time2State: An Unsupervised Framework for Inferring the Latent States in Time Series Data
Wang, C., Wu, K., Zhou, T., & Cai, Z. (2023) · Proceedings of the ACM on Management of Data · doi:10.1145/3588697 · Cited by 36
“There are dozens of time series representation learning methods [10, 12, 13, 22, 31, 33, 40], but very few consider segmentation as a downstream task.”
BackgroundIntelligent 3D garment system of the human body based on deep spiking neural network
Jiang, M., Tian, Z., Yu, C., Shi, Y., Liu, L., Peng, T., Hu, X., & Yu, F. Q. (2024) · Virtual Reality & Intelligent Hardware · doi:10.1016/j.vrih.2023.07.002 · Cited by 20
FreqMAE: Frequency-Aware Masked Autoencoder for Multi-Modal IoT Sensing
Kara, D., Kimura, T., Liu, S., Li, J., Liu, D., Wang, T., Wang, R., Chen, Y., Hu, Y., & Abdelzaher, T. (2024) · International World Wide Web Conference (WWW) · doi:10.1145/3589334.3645346 · Cited by 38
“…baselines: a supervised benchmark, five self-supervised frameworks (MAE [16], SimCLR [7], CAV-MAE [14], AudioMAE [19], LIMU-BERT [53]), two modality-matching contrastive baselines (CMC [ 42], Cosmo [30]), and two state-of-the-art (SOTA) contrastive time series frameworks (TS-TCC [11], TS2Vec [55]).”
BackgroundA Shapelet-Based Framework for Unsupervised Multivariate Time Series Representation Learning
Liang, Z., Zhang, J., Liang, C., Wang, H., Liang, Z., & Pan, L. (2023) · Proceedings of the VLDB Endowment · doi:10.14778/3632093.3632103 · Cited by 33
“Despite the success of contrastive learning in URL [9, 55, 58, 63], an open question is how to determine proper data augmentation”
BackgroundMethodInfluentialMa, P., Yang, X., Hu, W., Fu, T., & Zhou, C. (2024) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2024.109652 · Cited by 21
Mgformer: Multi-group transformer for multivariate time series classification
Wen, J., Zhang, N., Lu, X., Hu, Z., & Huang, H. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.108633 · Cited by 30
Cao, C., Yang, S., Li, M., & Li, C. (2023) · BMC Bioinformatics · doi:10.1186/s12859-023-05352-7 · Cited by 23
“It has been successfully applied to various fields such as NLP and CV, and many researchers[29-31] have incorporated Transformer as a sub-model and achieved impressive success.”
MethodTimeDRL: Disentangled Representation Learning for Multivariate Time-Series
Chang, C., Chan, C.-T., Wang, W., Peng, W.-C., & Chen, T.-F. (2024) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde60146.2024.00054 · Cited by 56
“Transformers have achieved notable success in downstream time-series tasks [35]–[38], but in the realm of self-supervised learning for time-series, CNN-based [12], [24] and RNN-based [14] models are generally preferred over Transformers.”
BackgroundMethodInfluentialSimilarity learning for wells based on logging data
Romanenkova, E., Rogulina, A., Shakirov, А., Stulov, N., Zaytsev, A., Ismailova, L., Kovalev, D., Katterbauer, K., & Alshehri, A. A. (2022) · Journal of Petroleum Science and Engineering · doi:10.1016/j.petrol.2022.110690 · Cited by 24
Approximating dynamic time warping with a convolutional neural network on EEG data
Lerogeron, H., Picot-Clémente, R., Rakotomamonjy, A., et al. (2023) · Pattern Recognition Letters · doi:10.1016/j.patrec.2023.05.012 · Cited by 20
Unsupervised Clustering of Microseismic Signals Using a Contrastive Learning Model
Yang, Z., Li, H., Tuo, X., Li, L., & Wen, J. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3240728 · Cited by 31
“tation from augmented data [38], [39], [40].”
BackgroundSun, Z., Wang, Z., & Jiang, M. (2024) · Talanta · doi:10.1016/j.talanta.2024.126076 · Cited by 20
Unsupervised Multivariate Time-Series Transformers for Seizure Identification on EEG
Potter, İ. Y., Zerveas, G., Eickhoff, C., & Duncan, D. (2022) · International Conference on Machine Learning and Applications · doi:10.1109/icmla55696.2022.00208 · Cited by 29
“Section II), focusing mostly on supervised machine learning methods using both manual feature extraction [7]–[10], as well as deep neural networks (DNNs) without manual feature extraction [11]–[13].”
LPClass: Lightweight Personalized Sensor Data Classification in Computational Social Systems
Yu, Q., & Sun, L. (2022) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2022.3156792 · Cited by 22
CRBPSA: CircRNA-RBP interaction sites identification using sequence structural attention model
Cao, C., Wang, C., Dai, Q., Zou, Q., & Wang, T. (2024) · BMC Biology · doi:10.1186/s12915-024-02055-0 · Cited by 23
Clustering-Based Contrastive Learning for Fault Diagnosis With Few Labeled Samples
Dai, Y., Mei, Z., Li, J., Li, Z., Wei, K., Ding, M., Guo, S., & Chen, W. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3346494 · Cited by 27
“Among different SSL methods, contrastive learning (CL) is a powerful representation learning approach, which can use the structure and intrinsic information of data to train neural networks and learn invariant features from augmented data [19], [20], [21].”
mulEEG: A Multi-view Representation Learning on EEG Signals
Kumar, V., Reddy, L., Sharma, S. K., Dadi, K., Yarra, C., Bapi, R. S., & Rajendran, S. P. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-16437-8_38 · Cited by 39
Zheng, H., Liu, H., Liu, Z., & Tan, J. (2024) · Advanced Engineering Informatics · doi:10.1016/j.aei.2024.102432 · Cited by 24
Unraveling the ‘Anomaly’ in Time Series Anomaly Detection: A Self-supervised Tri-domain Solution
Sun, Y., Pang, G., Ye, G., Chen, T., Hu, X., & Yin, H. (2024) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde60146.2024.00080 · Cited by 48
“Previous works in time series have employed various methods to select positive pairs, such as choosing adjacent segments [31] or augmented segments [8].”
BackgroundMethodTS-DP: An Efficient Data Processing Algorithm for Distribution Digital Twin Grid for Industry 5.0
Xiao, L., Han, D., Yang, C., Cai, J., Liang, W., & Li, K. (2023) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2023.3332099 · Cited by 21
“The following methods are used for comparison in data classification experiments: TS2Vec [40], T-Loss [49], TNC [50], TSTCC [51], DTW [52], TST [53].”
MethodModality Consistency-Guided Contrastive Learning for Wearable-Based Human Activity Recognition
Guo, C., Zhang, Y., Chen, Y., Xu, C., & Wang, Z. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3379019 · Cited by 24
“We compare the proposed ModCL method with five base-lines: 1) TF-C [40]; 2) TS2vec [15]; 3) TS-TCC [16]; 4) SimCLR [32]; and 5) mixing-up [42].”
BackgroundMethodLee, H., Seong, E., & Chae, D.-K. (2022) · Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2022/537 · Cited by 18
CutAddPaste: Time Series Anomaly Detection by Exploiting Abnormal Knowledge
Wang, R., Mou, X., Yang, R., Gao, K., Liu, P., Liu, C., Wo, T., & Xu-dong, L. (2024) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3637528.3671739 · Cited by 39
“Then, three single and three multiple deep normality assumptions-based AD methods are compared: LSTM Encoder-decoder (LSTM-ED) [1], Deep one-class (Deep SVDD) [2], Anomaly Transformer [33], COCA [3], AOC [17], and TCC [16, 34].”
TS-MAE: A masked autoencoder for time series representation learning
Liu, Q., Ye, J., Liang, H., Sun, L., & Du, B. (2024) · Information Sciences · doi:10.1016/j.ins.2024.121576 · Cited by 20
If only we had more data!: Sensor-Based Human Activity Recognition in Challenging Scenarios
Plötz, T. (2023) · 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops) · doi:10.1109/percomworkshops56833.2023.10150267 · Cited by 16
“Many methods have been developed for self-supervised representation learning on time-series data, including contrastive learning of augmented versions of input [38], and distinguishing signals from a close neighborhood to ones temporally farther away [39].”
MethodChen, X., Zhao, L., Xu, J., Liu, Z., Zhang, C., Xu, L., & Guo, N. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3341134 · Cited by 20
“Recently, contrastive representation learning has rapidly developed into one of the most representative self-supervised methods [39], [40], [41].”
BackgroundZhang, H., Chan, S., Qin, S., Dong, Z., & Chen, G. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112369 · Cited by 24
Basheer, S., Aldehim, G., Alluhaidan, A. S., & Sakri, S. B. (2024) · Alexandria Engineering Journal · doi:10.1016/j.aej.2024.06.058 · Cited by 19
Xia, S., Chu, L., Pei, L., Yang, J., Yu, W., & Qiu, R. C. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3381171 · Cited by 23
“[48]proposedacon-textual contrast module with weak and strong augmentations to Authorized licensed use limited to the terms of the applicable license agreement with IEEE.”
BackgroundGorade, V., Singh, A., & Mishra, D. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3331506 · Cited by 22
Wang, X., Yin, K., Ouyang, Q., Wen, X., Zhang, S., Zhang, W., Cao, L., Han, J., Jin, X., & Pei, D. (2022) · Proceedings/Proceedings - International Symposium on Software Reliability Engineering · doi:10.1109/issre55969.2022.00043 · Cited by 20
“These works typically apply data augmentation technique to generate time series segment pairs to train the model for various downstream tasks, i.e., time series classification in SimCLR-TS [63] and TimeCLR [65], representation learning in TS-TCC [64], change point detection in TS-CP 2 [41], and anomaly detection in Tim…”
MethodIntricate Spatiotemporal Dependency Learning for Temporal Knowledge Graph Reasoning
Li, X., Zhou, H., Yao, W., Li, W., Liu, B.-J., & Lin, Y. (2024) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3648366 · Cited by 16
“Recently, there are very few works trying to apply contrastive learning for capturing temporal dependencies of time-series data [ 45 ].”
BackgroundMethodInfluentialSelf-citationContrasting Estimation of Pattern Prototypes for Anomaly Detection in Urban Crowd Flow
Wang, Y., Luo, X., & Zhou, Z. (2024) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2024.3355143 · Cited by 15
“Contrastive learning, a popular approach in deep learning, facilitates unsupervised representation learning by repelling negative pairs while attracting positive pairs [21], [22], [23], [40].”
BackgroundMethodInfluentialWei, Y., Lian, C., Xu, B., Zhao, P., Yang, H., & Zeng, Z. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.111311 · Cited by 17
Unsupervised Domain-Agnostic Fake News Detection Using Multi-Modal Weak Signals
Silva, A., Luo, L., Karunasekera, S. A., & Leckie, C. A. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3392788 · Cited by 23
“To learn z p , we adopt the contrastive learning approach proposed in [36] that learns a representation for time-series data by maximizing the similarity among different correlated views of the same sample, while minimizing their similarity with other samples.”
MethodInfluentialEEG foundation models: a critical review of current progress and future directions
Kuruppu, G., Wagh, N., Křemen, V., & Varatharajah, Y. (2026) · Journal of Neural Engineering · doi:10.1088/1741-2552/ae4455 · Cited by 65
“The self-supervised pretrained baselines considered in this study included (RP, TS, CPC) [6], BENDR [124], MVTS [129], and BrainBERT that are designed for brain signals, and CoST [130], TF-C [131], PatchTST [132], TS-TCC [133] that are designed for general time series.”
Metadata
- Authors
- Eldele, E., Ragab, M., Chen, Z., Wu, M., Kwoh, C. K., Li, X., & Guan, C.
- Affiliations
- Nanyang Technological University (SG)
- Agency for Science, Technology and Research (SG)
- Venue
- Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence
- Published
- 2021-08
- Pages
- 2352-2359
- OpenAlex
- W3190152617
- Cited by
- 696
- References
- 25
- Citations by year
- 2026: 152 · 2025: 213 · 2024: 181 · 2023: 117 · 2022: 31 · 2021: 2
- FWCI
- 50.23
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- International Joint Conference on Artificial Intelligence (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- DR-NTU (Nanyang Technological University) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Funders
- Agency for Science, Technology and Research (A20H6b0151, C210112046)
- Topics
- Physical Sciences › Computer Science › Signal Processing › Time Series Analysis and Forecasting
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Advanced Graph Neural Networks
- Keywords
- time series representation learning, unsupervised learning, contrastive learning, data augmentation, temporal dynamics, limited labeled data, transfer learning
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.