Beyond Accuracy: Are Time Series Foundation Models Well-Calibrated?
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910267899641856 |
|---|---|
| author | Adler, Coen Chang, Yuxin Draxler, Felix Abdi, Samar Smyth, Padhraic |
| author_facet | Adler, Coen Chang, Yuxin Draxler, Felix Abdi, Samar Smyth, Padhraic |
| contents | The recent development of foundation models for time series data has generated considerable interest in using such models across a variety of applications. Although foundation models achieve state-of-the-art predictive performance, their calibration properties remain relatively underexplored, despite the fact that calibration can be critical for many practical applications. In this paper, we investigate the calibration-related properties of five recent time series foundation models and two competitive baselines. We perform a series of systematic evaluations assessing model calibration (i.e., over- or under-confidence), effects of varying prediction heads, and calibration under long-term autoregressive forecasting. We find that time series foundation models are consistently better calibrated than baseline models and tend not to be either systematically over- or under-confident, in contrast to the overconfidence often seen in other deep learning models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_16060 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Accuracy: Are Time Series Foundation Models Well-Calibrated? Adler, Coen Chang, Yuxin Draxler, Felix Abdi, Samar Smyth, Padhraic Machine Learning Artificial Intelligence Methodology The recent development of foundation models for time series data has generated considerable interest in using such models across a variety of applications. Although foundation models achieve state-of-the-art predictive performance, their calibration properties remain relatively underexplored, despite the fact that calibration can be critical for many practical applications. In this paper, we investigate the calibration-related properties of five recent time series foundation models and two competitive baselines. We perform a series of systematic evaluations assessing model calibration (i.e., over- or under-confidence), effects of varying prediction heads, and calibration under long-term autoregressive forecasting. We find that time series foundation models are consistently better calibrated than baseline models and tend not to be either systematically over- or under-confident, in contrast to the overconfidence often seen in other deep learning models. |
| title | Beyond Accuracy: Are Time Series Foundation Models Well-Calibrated? |
| topic | Machine Learning Artificial Intelligence Methodology |
| url | https://arxiv.org/abs/2510.16060 |