Probabilistic conformal prediction with approximate conditional validity
Plassier, Vincent Fishkov, Alexander Guizani, Mohsen Panov, Maxim Moulines, Eric
Plassier, Vincent
Fishkov, Alexander
Guizani, Mohsen
Panov, Maxim
Moulines, Eric
Supervisor
Department
Machine Learning
Embargo End Date
Type
Conference proceeding
Date
2025
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
We develop a new method for generating prediction sets that combines the flexibility of conformal methods with an estimate of the conditional distribution PY |X. Existing methods, such as conformalized quantile regression and probabilistic conformal prediction, usually provide only a marginal coverage guarantee. In contrast, our approach extends these frameworks to achieve approximate conditional coverage, which is crucial for many practical applications. Our prediction sets adapt to the behavior of the predictive distribution, making them effective even under high heteroscedasticity. While exact conditional guarantees are infeasible without assumptions on the underlying data distribution, we derive non-asymptotic bounds that depend on the total variation distance between the conditional distribution and its estimate. Using extensive simulations, we show that our method consistently outperforms existing approaches in terms of conditional coverage, leading to more reliable statistical inference in a variety of applications. © 2025 13th International Conference on Learning Representations, ICLR 2025. All rights reserved.
Citation
V. Plassier, A. Fishkov, M. Guizani, M. Panov, and E. Moulines, “Probabilistic Conformal Prediction with Approximate Conditional Validity,” International Conference on Representation Learning, vol. 2025, pp. 62896–62928, May 2025
Source
13th International Conference on Learning Representations, ICLR 2025
Conference
13th International Conference on Learning Representations, ICLR 2025
Keywords
Subjects
Source
13th International Conference on Learning Representations, ICLR 2025
Publisher
International Conference on Learning Representations, ICLR