MHCCL: Masked Hierarchical Cluster-Wise Contrastive Learning for Multivariate Time Series
Open accessAbstract
Learning semantic-rich representations from raw unlabeled time series data is critical for downstream tasks such as classification and forecasting. Contrastive learning has recently shown its promising representation learning capability in the absence of expert annotations. However, existing contrastive approaches generally treat each instance independently, which leads to false negative pairs that share the same semantics. To tackle this problem, we propose MHCCL, a Masked Hierarchical Cluster-wise Contrastive Learning model, which exploits semantic information obtained from the hierarchical structure consisting of multiple latent partitions for multivariate time series. Motivated by the observation that fine-grained clustering preserves higher purity while coarse-grained one reflects higher-level semantics, we propose a novel downward masking strategy to filter out fake negatives and supplement positives by incorporating the multi-granularity information from the clustering hierarchy. In addition, a novel upward masking strategy is designed in MHCCL to remove outliers of clusters at each partition to refine prototypes, which helps speed up the hierarchical clustering process and improves the clustering quality. We conduct experimental evaluations on seven widely-used multivariate time series datasets. The results demonstrate the superiority of MHCCL over the state-of-the-art approaches for unsupervised time series representation learning.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
In the upward mask strategy, MHCCL believes that outliers greatly impact prototypes, so these outliers should be removed when updating prototypes.
“In the upward mask strategy, MHCCL believes that outliers greatly impact prototypes, so these outliers should be removed when updating prototypes.”
From Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects (2024) · cites this paper as [139]
MHCCL proposes a hierarchical clustering based on the upward masking strategy and a contrastive pairs selection strategy based on the downward masking strategy.
“MHCCL [139] proposes a hierarchical clustering based on the upward masking strategy and a contrastive pairs selection strategy based on the downward masking strategy.”
From Self-Supervised Learning for Time Series Analysis: Taxonomy, Progress, and Prospects (2024) · cites this paper as [139]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
Activity recognition using cell phone accelerometers
Kwapisz, J. R., Weiss, G. M., & Moore, S. A. (2011) · ACM SIGKDD Explorations Newsletter · doi:10.1145/1964897.1964918 · Cited by 2,912 · see citing works
Bagnall, A., Lines, J., Bostrom, A., Large, J., & Keogh, E. (2017) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-016-0483-9 · Cited by 1,377 · see citing works
The UEA multivariate time series classification archive, 2018
Bagnall, A., Dau, H. A., Lines, J., Flynn, M., Large, J., Bostrom, A., Southam, P., & Keogh, E. (2018) · arXiv · doi:10.48550/arxiv.1811.00075 · Cited by 87 · see citing works
What 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 · see citing works
Self-citationUniMiB SHAR: a new dataset for human activity recognition using acceleration data from smartphonesPreprint
Micucci, D., Mobilio, M., & Napoletano, P. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.07688 · Cited by 30 · see citing works
Deldari, S., Xue, H., Saeed, A., He, J., Smith, D. V., & Salim, F. D. (2022) · arXiv · doi:10.48550/arxiv.2206.02353 · Cited by 26 · see citing works
Improving Clinical Predictions through Unsupervised Time Series Representation LearningPreprint
Lyu, X., Hüser, M., Hyland, S. L., Zerveas, G., & Rätsch, G. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.00490 · Cited by 22 · see citing works
SLIC: Self-Supervised Learning with Iterative Clustering for Human Action VideosPreprint
Khorasgani, S. H., Chen, Y., & Shkurti, F. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2206.12534
Flagged first, then by citations
Research works printed but not linked: 23
- Andrzejak, R. G.; Lehnertz, K.; Mormann, F.; Rieke, C.; David, P.; and Elger, C. E. 2001. Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state. Physical Review E, 64(6): 061907.
- Belkhouja, T.; Yan, Y.; and Doppa, J. R. 2022. Training Robust Deep Models for Time-Series Domain: Novel Al- gorithms and Theoretical Analysis. In AAAI, 6055-6063. AAAI Press.
- Caron, M.; Misra, I.; Mairal, J.; Goyal, P.; Bojanowski, P.; and Joulin, A. 2020. Unsupervised Learning of Visual Fea- tures by Contrasting Cluster Assignments. In NeurIPS.
- Chen, P.; Huang, D.; He, D.; Long, X.; Zeng, R.; Wen, S.; Tan, M.; and Gan, C. 2021. RSPNet: Relative Speed Per- ception for Unsupervised Video Representation Learning. In AAAI, 1045-1053. AAAI Press.
- Chen, T.; Kornblith, S.; Norouzi, M.; and Hinton, G. E. 2020. A Simple Framework for Contrastive Learning of Vi- sual Representations. In ICML, volume 119 of Proceedings of Machine Learning Research, 1597-1607.
- PMLR.
- Eldele, E.; Ragab, M.; Chen, Z.; Wu, M.; Kwoh, C. K.; Li, X.; and Guan, C. 2021. Time-Series Representation Learn- ing via Temporal and Contextual Contrasting. In IJCAI, 2352-2359. ijcai.org.
- Franceschi, J.; Dieuleveut, A.; and Jaggi, M. 2019. Unsu- pervised Scalable Representation Learning for Multivariate Time Series. In NeurIPS, 4652-4663.
- Grill, J.; Strub, F.; Altché, F.; Tallec, C.; Richemond, P. H.; Buchatskaya, E.; Doersch, C.; Pires, B. Á.; Guo, Z.; Azar, M. G.; Piot, B.; Kavukcuoglu, K.; Munos, R.; and Valko, M. 2020. Bootstrap Your Own Latent -A New Approach to Self-Supervised Learning. In NeurIPS.
- He, K.; Fan, H.; Wu, Y.; Xie, S.; and Girshick, R. B. 2020. Momentum Contrast for Unsupervised Visual Representa- tion Learning. In CVPR, 9726-9735. Computer Vision Foundation / IEEE.
- Kolesnikov, A.; Zhai, X.; and Beyer, L. 2019. Revisiting Self-Supervised Visual Representation Learning. In CVPR, 1920-1929. Computer Vision Foundation / IEEE.
- Li, J.; Zhou, P.; Xiong, C.; and Hoi, S. C. H. 2021a. Pro- totypical Contrastive Learning of Unsupervised Representa- tions. In ICLR. OpenReview.net.
- Li, Y.; Hu, P.; Liu, J. Z.; Peng, D.; Zhou, J. T.; and Peng, X. 2021b. Contrastive Clustering. In AAAI, 8547-8555. AAAI Press.
- Misra, I.; and van der Maaten, L. 2020. Self-Supervised Learning of Pretext-Invariant Representations. In CVPR, 6706-6716. Computer Vision Foundation / IEEE.
- Rambhatla, S.; Che, Z.; and Liu, Y. 2022. I-SEA: Impor- tance Sampling and Expected Alignment-Based Deep Dis- tance Metric Learning for Time Series Analysis and Embed- ding. In AAAI, 8045-8053. AAAI Press.
- Ruiz, A. P.; Flynn, M.; Large, J.; Middlehurst, M.; and Bag- nall, A. J. 2021. The great multivariate time series classifi- cation bake off: a review and experimental evaluation of re- cent algorithmic advances. Data Min. Knowl. Discov., 35(2): 401-449.
- Sarfraz, M. S.; Sharma, V.; and Stiefelhagen, R. 2019. Effi- cient Parameter-Free Clustering Using First Neighbor Rela- tions. In CVPR, 8934-8943. Computer Vision Foundation / IEEE.
- Sarkar, P.; and Etemad, A. 2020. Self-supervised ECG Representation Learning for Emotion Recognition. CoRR, abs/2002.03898. · arXiv
- Sharma, V.; Tapaswi, M.; Sarfraz, M. S.; and Stiefelhagen, R. 2020. Clustering based Contrastive Learning for Improv- ing Face Representations. In FG, 109-116. IEEE.
- Tonekaboni, S.; Eytan, D.; and Goldenberg, A. 2021. Unsu- pervised Representation Learning for Time Series with Tem- poral Neighborhood Coding. In ICLR. OpenReview.net.
- Wang, X.; Liu, Z.; and Yu, S. X. 2021. Unsupervised Feature Learning by Cross-Level Instance-Group Discrimination. In CVPR, 12586-12595. Computer Vision Foundation / IEEE.
- Yue, Z.; Wang, Y.; Duan, J.; Yang, T.; Huang, C.; Tong, Y.; and Xu, B. 2022. TS2Vec: Towards Universal Representa- tion of Time Series. In AAAI, 8980-8987. AAAI Press.
- Zhang, Y.; Li, Y.; Zhou, X.; Liu, Z.; and Luo, J. 2021. C 3 - GAN: Complex-Condition-Controlled Urban Traffic Esti- mation through Generative Adversarial Networks. In ICDM, 1505-1510. IEEE.
Cited by
Sorted by citations
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 · see citing works
Self-citationFully-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 · see citing works
Self-citationComparing self-supervised learning techniques for wearable human activity recognition
Ek, S., Presotto, R., Civitarese, G., Portet, F., Lalanda, P., & Bettini, C. (2025) · CCF Transactions on Pervasive Computing and Interaction · doi:10.1007/s42486-024-00182-9 · Cited by 15 · see citing works
- Authors
- Meng, Q., Qian, H., Liu, Y., Cui, L., Xu, Y., & Shen, Z.
- Affiliations
- Shandong University (CN)
- Lund University (SE)
- Nanyang Technological University (SG)
- Venue
- Proceedings of the AAAI Conference on Artificial Intelligence
- Published
- 2023-06-26
- Volume
- 37
- Issue
- 8
- Pages
- 9153-9161
- ISSN
- 2159-5399 (ISSN-L)
- 2374-3468
- OpenAlex
- W4382317959
- Cited by
- 52
- References
- 9
- Citations by year
- 2026: 10 · 2025: 22 · 2024: 19 · 2023: 1
- FWCI
- 10.82
- Type
- conference paper
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Proceedings of the AAAI Conference on Artificial Intelligence (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Lund University Publications (Lund University) (submittedVersion)
- AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Funders
- Shandong University
- National Natural Science Foundation of China (91846205)
- China Scholarship Council
- Natural Science Foundation of Shandong Province (91846205, ZR202111180007)
- Topics
- Physical Sciences › Computer Science › Signal Processing › Time Series Analysis and Forecasting
- Physical Sciences › Computer Science › Artificial Intelligence › Advanced Clustering Algorithms Research
- Physical Sciences › Computer Science › Computational Theory and Mathematics › Topological and Geometric Data Analysis
- Keywords
- contrastive learning, multivariate time series, unsupervised feature learning, hierarchical clustering, prototype refinement, time series classification, time series forecasting
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
