Evaluating Bayesian deep learning for radio galaxy classification

TL;DR

This paper assesses various Bayesian neural networks for radio galaxy classification, focusing on their predictive accuracy, uncertainty calibration, and ability to detect distribution shifts in the context of upcoming radio astronomy data.

Contribution

It provides a comparative evaluation of BNNs in radio astronomy, highlighting their strengths and limitations for uncertainty estimation and model robustness.

Findings

BNNs improve uncertainty calibration over standard neural networks

Certain BNN architectures excel in distribution-shift detection

Bayesian methods enhance the reliability of radio galaxy classification

Abstract

The radio astronomy community is rapidly adopting deep learning techniques to deal with the huge data volumes expected from the next generation of radio observatories. Bayesian neural networks (BNNs) provide a principled way to model uncertainty in the predictions made by such deep learning models and will play an important role in extracting well-calibrated uncertainty estimates on their outputs. In this work, we evaluate the performance of different BNNs against the following criteria: predictive performance, uncertainty calibration and distribution-shift detection for the radio galaxy classification problem.