Leveraging Generic Time Series Foundation Models for EEG Classification
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914127025274880 |
|---|---|
| author | Gnassounou, Théo Moakher, Yessin Xie, Shifeng Feofanov, Vasilii Redko, Ievgen |
| author_facet | Gnassounou, Théo Moakher, Yessin Xie, Shifeng Feofanov, Vasilii Redko, Ievgen |
| contents | Foundation models for time series are emerging as powerful general-purpose backbones, yet their potential for domain-specific biomedical signals such as electroencephalography (EEG) remains rather unexplored. In this work, we investigate the applicability a recently proposed time series classification foundation model, to a different EEG tasks such as motor imagery classification and sleep stage prediction. We test two pretraining regimes: (a) pretraining on heterogeneous real-world time series from multiple domains, and (b) pretraining on purely synthetic data. We find that both variants yield strong performance, consistently outperforming EEGNet, a widely used convolutional baseline, and CBraMod, the most recent EEG-specific foundation model. These results suggest that generalist time series foundation models, even when pretrained on data of non-neural origin or on synthetic signals, can transfer effectively to EEG. Our findings highlight the promise of leveraging cross-domain pretrained models for brain signal analysis, suggesting that EEG may benefit from advances in the broader time series literature. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_27522 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Leveraging Generic Time Series Foundation Models for EEG Classification Gnassounou, Théo Moakher, Yessin Xie, Shifeng Feofanov, Vasilii Redko, Ievgen Machine Learning Artificial Intelligence Foundation models for time series are emerging as powerful general-purpose backbones, yet their potential for domain-specific biomedical signals such as electroencephalography (EEG) remains rather unexplored. In this work, we investigate the applicability a recently proposed time series classification foundation model, to a different EEG tasks such as motor imagery classification and sleep stage prediction. We test two pretraining regimes: (a) pretraining on heterogeneous real-world time series from multiple domains, and (b) pretraining on purely synthetic data. We find that both variants yield strong performance, consistently outperforming EEGNet, a widely used convolutional baseline, and CBraMod, the most recent EEG-specific foundation model. These results suggest that generalist time series foundation models, even when pretrained on data of non-neural origin or on synthetic signals, can transfer effectively to EEG. Our findings highlight the promise of leveraging cross-domain pretrained models for brain signal analysis, suggesting that EEG may benefit from advances in the broader time series literature. |
| title | Leveraging Generic Time Series Foundation Models for EEG Classification |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2510.27522 |