Mirror Online Conformal Prediction with Intermittent Feedback
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918084510482432 |
|---|---|
| author | Wang, Bowen Zecchin, Matteo Simeone, Osvaldo |
| author_facet | Wang, Bowen Zecchin, Matteo Simeone, Osvaldo |
| contents | Online conformal prediction enables the runtime calibration of a pre-trained artificial intelligence model using feedback on its performance. Calibration is achieved through set predictions that are updated via online rules so as to ensure long-term coverage guarantees. While recent research has demonstrated the benefits of incorporating prior knowledge into the calibration process, this has come at the cost of replacing coverage guarantees with less tangible regret guarantees based on the quantile loss. This work introduces intermittent mirror online conformal prediction (IM-OCP), a novel runtime calibration framework that integrates prior knowledge, operates under potentially intermittent feedback, and features minimal memory complexity. IM-OCP guarantees long-term coverage and sub-linear regret, both of which hold deterministically for any given data sequence and in expectation with respect to the intermittent feedback. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_10345 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mirror Online Conformal Prediction with Intermittent Feedback Wang, Bowen Zecchin, Matteo Simeone, Osvaldo Machine Learning Signal Processing Online conformal prediction enables the runtime calibration of a pre-trained artificial intelligence model using feedback on its performance. Calibration is achieved through set predictions that are updated via online rules so as to ensure long-term coverage guarantees. While recent research has demonstrated the benefits of incorporating prior knowledge into the calibration process, this has come at the cost of replacing coverage guarantees with less tangible regret guarantees based on the quantile loss. This work introduces intermittent mirror online conformal prediction (IM-OCP), a novel runtime calibration framework that integrates prior knowledge, operates under potentially intermittent feedback, and features minimal memory complexity. IM-OCP guarantees long-term coverage and sub-linear regret, both of which hold deterministically for any given data sequence and in expectation with respect to the intermittent feedback. |
| title | Mirror Online Conformal Prediction with Intermittent Feedback |
| topic | Machine Learning Signal Processing |
| url | https://arxiv.org/abs/2503.10345 |