Minimax Multi-Target Conformal Prediction with Applications to Imaging Inverse Problems

TL;DR

This paper introduces a minimax conformal prediction method for multiple targets in ill-posed imaging inverse problems, ensuring reliable uncertainty quantification with applications demonstrated on MRI data.

Contribution

It proposes a novel asymptotic minimax approach for multi-target conformal prediction that guarantees joint coverage and tight prediction intervals, extending existing scalar-target methods.

Findings

Outperforms existing multi-target conformal prediction methods in experiments.

Provides tight prediction intervals with guaranteed joint marginal coverage.

Demonstrates effectiveness on synthetic and MRI imaging data.

Abstract

In ill-posed imaging inverse problems, uncertainty quantification remains a fundamental challenge, especially in safety-critical applications. Recently, conformal prediction has been used to quantify the uncertainty that the inverse problem contributes to downstream tasks like image classification, image quality assessment, fat mass quantification, etc. While existing works handle only a scalar estimation target, practical applications often involve multiple targets. In response, we propose an asymptotically minimax approach to multi-target conformal prediction that provides tight prediction intervals while ensuring joint marginal coverage. We then outline how our minimax approach can be applied to multi-metric blind image quality assessment, multi-task uncertainty quantification, and multi-round measurement acquisition. Finally, we numerically demonstrate the benefits of our minimax method, relative to existing multi-target conformal prediction methods, using both synthetic and magnetic resonance imaging (MRI) data. Code is available at https://github.com/jwen307/multi_target_minimax.