Efficient Conformal Prediction under Data Heterogeneity
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914868448198656 |
|---|---|
| author | Plassier, Vincent Kotelevskii, Nikita Rubashevskii, Aleksandr Noskov, Fedor Velikanov, Maksim Fishkov, Alexander Horvath, Samuel Takac, Martin Moulines, Eric Panov, Maxim |
| author_facet | Plassier, Vincent Kotelevskii, Nikita Rubashevskii, Aleksandr Noskov, Fedor Velikanov, Maksim Fishkov, Alexander Horvath, Samuel Takac, Martin Moulines, Eric Panov, Maxim |
| contents | Conformal Prediction (CP) stands out as a robust framework for uncertainty quantification, which is crucial for ensuring the reliability of predictions. However, common CP methods heavily rely on data exchangeability, a condition often violated in practice. Existing approaches for tackling non-exchangeability lead to methods that are not computable beyond the simplest examples. This work introduces a new efficient approach to CP that produces provably valid confidence sets for fairly general non-exchangeable data distributions. We illustrate the general theory with applications to the challenging setting of federated learning under data heterogeneity between agents. Our method allows constructing provably valid personalized prediction sets for agents in a fully federated way. The effectiveness of the proposed method is demonstrated in a series of experiments on real-world datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_15799 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Efficient Conformal Prediction under Data Heterogeneity Plassier, Vincent Kotelevskii, Nikita Rubashevskii, Aleksandr Noskov, Fedor Velikanov, Maksim Fishkov, Alexander Horvath, Samuel Takac, Martin Moulines, Eric Panov, Maxim Machine Learning Conformal Prediction (CP) stands out as a robust framework for uncertainty quantification, which is crucial for ensuring the reliability of predictions. However, common CP methods heavily rely on data exchangeability, a condition often violated in practice. Existing approaches for tackling non-exchangeability lead to methods that are not computable beyond the simplest examples. This work introduces a new efficient approach to CP that produces provably valid confidence sets for fairly general non-exchangeable data distributions. We illustrate the general theory with applications to the challenging setting of federated learning under data heterogeneity between agents. Our method allows constructing provably valid personalized prediction sets for agents in a fully federated way. The effectiveness of the proposed method is demonstrated in a series of experiments on real-world datasets. |
| title | Efficient Conformal Prediction under Data Heterogeneity |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2312.15799 |