Time Series Class-Incremental Learning via Confidence-guided Mask Distillation and Prototype-guided Contrastive Learning
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (28) · 23917-23925
DOI: 10.1609/aaai.v40i28.39568 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yu Liu und 6 weiteren Autor:innen.
Abstract (Originalsprache)
Class-incremental learning (CIL) has recently gained great attention in the field of time series classification. Existing CIL methods based on knowledge distillation exhibit impressive ability to retain prior knowledge and overcome catastrophic forgetting, however, their effectiveness faces major challenges posed by time series data. Since temporal data is more susceptible to sensor errors and electronic noise, the distillation process may be significantly affected by noisy knowledge transfer. To address this issue, we propose a novel confidence-guided mask distillation (CMD) framework, to prevent the noisy inheritance during distillation. The core of CMD lies in a dynamic masking mechanism guided by prediction confidence, capable of allocating higher weights to high-confidence time series and substantially suppressing the influence of low-confidence ones. Additionally, different from prior work simply passing a set of feature prototypes to the classifier, we develop prototype-guided contrastive learning (PCL) to alleviate the classifier bias on new classes, through extra contrastive constraints to push away the feature distributions of old feature prototypes from those of new classes features. Extensive experiments on three time-series datasets demonstrate that, our method significantly outperforms other replay-free CIL approaches in raising average accuracy, as well as decreasing forgetting rate.
Zitieren
APA
Liu, Y., Yang, H., Sui, J., Wang, H., Li, H., Wang, W., & Jia, Q. (2026). Time Series Class-Incremental Learning via Confidence-guided Mask Distillation and Prototype-guided Contrastive Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(28), 23917-23925. https://doi.org/10.1609/aaai.v40i28.39568
Vancouver
Liu Y, Yang H, Sui J, Wang H, Li H, Wang W, et al. Time Series Class-Incremental Learning via Confidence-guided Mask Distillation and Prototype-guided Contrastive Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(28):23917-23925. doi:10.1609/aaai.v40i28.39568
BibTeX
@article{Liu2026,
author = {Liu, Yu and Yang, Haoqin and Sui, Jinping and Wang, Hui and Li, Haipeng and Wang, Weimin and Jia, Qi},
title = {Time Series Class-Incremental Learning via Confidence-guided Mask Distillation and Prototype-guided Contrastive Learning},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {28},
pages = {23917--23925},
doi = {10.1609/aaai.v40i28.39568}
}
EUVIMED