Bayesian Imaging for Radio Interferometry with Score-Based Priors

TL;DR

This paper introduces a Bayesian imaging method using score-based priors from optical galaxy images to improve radio interferometry image reconstruction, providing plausible uncertainty estimates and competitive results.

Contribution

It presents a novel score-based prior approach for radio interferometry imaging that effectively incorporates optical galaxy data to enhance image quality and uncertainty quantification.

Findings

Produces plausible posterior samples despite prior misspecification

Achieves results competitive with existing radio interferometry algorithms

Demonstrates effective use of optical images as priors in radio imaging

Abstract

The inverse imaging task in radio interferometry is a key limiting factor to retrieving Bayesian uncertainties in radio astronomy in a computationally effective manner. We use a score-based prior derived from optical images of galaxies to recover images of protoplanetary disks from the DSHARP survey. We demonstrate that our method produces plausible posterior samples despite the misspecified galaxy prior. We show that our approach produces results which are competitive with existing radio interferometry imaging algorithms.