Mantis: Lightweight Calibrated Foundation Model for User-Friendly Time Series 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_ | 1866913702003867648 |
|---|---|
| author | Feofanov, Vasilii Wen, Songkang Alonso, Marius Ilbert, Romain Guo, Hongbo Tiomoko, Malik Pan, Lujia Zhang, Jianfeng Redko, Ievgen |
| author_facet | Feofanov, Vasilii Wen, Songkang Alonso, Marius Ilbert, Romain Guo, Hongbo Tiomoko, Malik Pan, Lujia Zhang, Jianfeng Redko, Ievgen |
| contents | In recent years, there has been increasing interest in developing foundation models for time series data that can generalize across diverse downstream tasks. While numerous forecasting-oriented foundation models have been introduced, there is a notable scarcity of models tailored for time series classification. To address this gap, we present Mantis, a new open-source foundation model for time series classification based on the Vision Transformer (ViT) architecture that has been pre-trained using a contrastive learning approach. Our experimental results show that Mantis outperforms existing foundation models both when the backbone is frozen and when fine-tuned, while achieving the lowest calibration error. In addition, we propose several adapters to handle the multivariate setting, reducing memory requirements and modeling channel interdependence. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_15637 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mantis: Lightweight Calibrated Foundation Model for User-Friendly Time Series Classification Feofanov, Vasilii Wen, Songkang Alonso, Marius Ilbert, Romain Guo, Hongbo Tiomoko, Malik Pan, Lujia Zhang, Jianfeng Redko, Ievgen Machine Learning Artificial Intelligence In recent years, there has been increasing interest in developing foundation models for time series data that can generalize across diverse downstream tasks. While numerous forecasting-oriented foundation models have been introduced, there is a notable scarcity of models tailored for time series classification. To address this gap, we present Mantis, a new open-source foundation model for time series classification based on the Vision Transformer (ViT) architecture that has been pre-trained using a contrastive learning approach. Our experimental results show that Mantis outperforms existing foundation models both when the backbone is frozen and when fine-tuned, while achieving the lowest calibration error. In addition, we propose several adapters to handle the multivariate setting, reducing memory requirements and modeling channel interdependence. |
| title | Mantis: Lightweight Calibrated Foundation Model for User-Friendly Time Series Classification |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2502.15637 |