Radio Galaxy Zoo: Using semi-supervised learning to leverage large unlabelled data-sets for radio galaxy classification under data-set shift

Abstract

In this work we examine the classification accuracy and robustness of a state-of-the-art semi-supervised learning (SSL) algorithm applied to the morphological classification of radio galaxies. We test if SSL with fewer labels can achieve test accuracies comparable to the supervised state-of-the-art and whether this holds when incorporating previously unseen data. We find that for the radio galaxy classification problem considered, SSL provides additional regularisation and outperforms the baseline test accuracy. However, in contrast to model performance metrics reported on computer science benchmarking data-sets, we find that improvement is limited to a narrow range of label volumes, with performance falling off rapidly at low label volumes. Additionally, we show that SSL does not improve model calibration, regardless of whether classification is improved. Moreover, we find that when different underlying catalogues drawn from the same radio survey are used to provide the labelled and unlabelled data-sets required for SSL, a significant drop in classification performance is observered, highlighting the difficulty of applying SSL techniques under dataset shift. We show that a class-imbalanced unlabelled data pool negatively affects performance through prior probability shift, which we suggest may explain this performance drop, and that using the Frechet Distance between labelled and unlabelled data-sets as a measure of data-set shift can provide a prediction of model performance, but that for typical radio galaxy data-sets with labelled sample volumes of O(1000), the sample variance associated with this technique is high and the technique is in general not sufficiently robust to replace a train-test cycle.

Accepted to MNRAS. 14 pages

References in corpus (17)

• The Rapid ASKAP Continuum Survey I: Design and First Results
• Revisiting the Fanaroff-Riley dichotomy and radio-galaxy morphology with the LOFAR Two-Metre Sky Survey (LoTSS)
• Deep ATLAS radio observations of the CDFS-SWIRE field
• A pilot study of the radio-emitting AGN population: the emerging new class of FR0 radio-galaxies
• Classifying Radio Galaxies with Convolutional Neural Network
• The Energetics and Lifetimes of Local Radio Active Galactic Nuclei
• A representative survey of the dynamics and energetics of FRII radio galaxies
• Cataloging the radio-sky with unsupervised machine learning: a new approach for the SKA era
• CNN Architecture Comparison for Radio Galaxy Classification
• Fanaroff-Riley classification of radio galaxies using group-equivariant convolutional neural networks
• Attention-gating for improved radio galaxy classification
• Morphological classification of compact and extended radio galaxies using convolutional neural networks and data augmentation techniques
• Rapid sorting of radio galaxy morphology using Haralick features
• Quantifying Uncertainty in Deep Learning Approaches to Radio Galaxy Classification
• Structured Variational Inference for Simulating Populations of Radio Galaxies
• Ram pressure statistics for bent tail radio galaxies
• RadioGAN - Translations between different radio surveys with generative adversarial networks