Zono-Conformal Prediction: Zonotope-Based Uncertainty Quantification for Regression and Classification Tasks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lützow, Laura, Eichelbeck, Michael, Kochenderfer, Mykel J., Althoff, Matthias
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915832585519104
author Lützow, Laura
Eichelbeck, Michael
Kochenderfer, Mykel J.
Althoff, Matthias
author_facet Lützow, Laura
Eichelbeck, Michael
Kochenderfer, Mykel J.
Althoff, Matthias
contents Conformal prediction is a popular uncertainty quantification method that augments a base predictor to return sets of predictions with statistically valid coverage guarantees. However, current methods are often computationally expensive and data-intensive, as they require constructing an uncertainty model before calibration. Moreover, existing approaches typically represent the prediction sets with intervals, which limits their ability to capture dependencies in multi-dimensional outputs. We address these limitations by introducing zono-conformal prediction, a novel approach inspired by interval predictor models and reachset-conformant identification that constructs prediction zonotopes with assured coverage. By placing zonotopic uncertainty sets directly into the model of the base predictor, zono-conformal predictors can be identified via a single, data-efficient linear program. While we can apply zono-conformal prediction to arbitrary nonlinear base predictors, we focus on feed-forward neural networks in this work. Aside from regression tasks, we also construct optimal zono-conformal predictors in classification settings where the output of an uncertain predictor is a set of possible classes. We provide probabilistic coverage guarantees and present methods for detecting outliers in the identification data. In extensive numerical experiments, we show that zono-conformal predictors are less conservative than interval predictor models and standard conformal prediction methods, while achieving a similar coverage over the test data.
format Preprint
id arxiv_https___arxiv_org_abs_2508_11025
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Zono-Conformal Prediction: Zonotope-Based Uncertainty Quantification for Regression and Classification Tasks
Lützow, Laura
Eichelbeck, Michael
Kochenderfer, Mykel J.
Althoff, Matthias
Machine Learning
Artificial Intelligence
Systems and Control
Conformal prediction is a popular uncertainty quantification method that augments a base predictor to return sets of predictions with statistically valid coverage guarantees. However, current methods are often computationally expensive and data-intensive, as they require constructing an uncertainty model before calibration. Moreover, existing approaches typically represent the prediction sets with intervals, which limits their ability to capture dependencies in multi-dimensional outputs. We address these limitations by introducing zono-conformal prediction, a novel approach inspired by interval predictor models and reachset-conformant identification that constructs prediction zonotopes with assured coverage. By placing zonotopic uncertainty sets directly into the model of the base predictor, zono-conformal predictors can be identified via a single, data-efficient linear program. While we can apply zono-conformal prediction to arbitrary nonlinear base predictors, we focus on feed-forward neural networks in this work. Aside from regression tasks, we also construct optimal zono-conformal predictors in classification settings where the output of an uncertain predictor is a set of possible classes. We provide probabilistic coverage guarantees and present methods for detecting outliers in the identification data. In extensive numerical experiments, we show that zono-conformal predictors are less conservative than interval predictor models and standard conformal prediction methods, while achieving a similar coverage over the test data.
title Zono-Conformal Prediction: Zonotope-Based Uncertainty Quantification for Regression and Classification Tasks
topic Machine Learning
Artificial Intelligence
Systems and Control
url https://arxiv.org/abs/2508.11025