Weight Pruning and Uncertainty in Radio Galaxy Classification

TL;DR

This paper uses variational inference to quantify uncertainty in radio galaxy classification, explores weight pruning and model calibration, and discusses the cold posterior effect's implications for Bayesian deep learning in astronomy.

Contribution

It demonstrates how Bayesian methods can quantify uncertainty, optimize model pruning, and investigates the cold posterior effect in radio galaxy classification.

Findings

Sparse priors improve uncertainty calibration.

30% pruning of fully-connected layers is possible without performance loss.

The cold posterior effect is linked to likelihood misspecification.

Abstract

In this work we use variational inference to quantify the degree of epistemic uncertainty in model predictions of radio galaxy classification and show that the level of model posterior variance for individual test samples is correlated with human uncertainty when labelling radio galaxies. We explore the model performance and uncertainty calibration for a variety of different weight priors and suggest that a sparse prior produces more well-calibrated uncertainty estimates. Using the posterior distributions for individual weights, we show that signal-to-noise ratio (SNR) ranking allows pruning of the fully-connected layers to the level of 30% without significant loss of performance, and that this pruning increases the predictive uncertainty in the model. Finally we show that, like other work in this field, we experience a cold posterior effect. We examine whether adapting the cost function in our model to accommodate model misspecification can compensate for this effect, but find that it does not make a significant difference. We also examine the effect of principled data augmentation and find that it improves upon the baseline but does not compensate for the observed effect fully. We interpret this as the cold posterior effect being due to the overly effective curation of our training sample leading to likelihood misspecification, and raise this as a potential issue for Bayesian deep learning approaches to radio galaxy classification in future.