CzSL: Learning from citizen science, experts and unlabelled data in astronomical image classification

TL;DR

This paper introduces a novel machine learning framework that leverages citizen science, expert, and unlabelled data for improved astronomical image classification, specifically for galaxy images, by combining autoencoders and transfer learning.

Contribution

It proposes an innovative unified learning approach that integrates unlabelled, amateur-labelled, and expert-labelled data using autoencoders and transfer learning for galaxy classification.

Findings

Improved classification accuracy over baseline methods.

Effective handling of imbalanced and multi-class scenarios.

Demonstrated robustness across different label confidence levels.

Abstract

Citizen science is gaining popularity as a valuable tool for labelling large collections of astronomical images by the general public. This is often achieved at the cost of poorer quality classifications made by amateur participants, which are usually verified by employing smaller data sets labelled by professional astronomers. Despite its success, citizen science alone will not be able to handle the classification of current and upcoming surveys. To alleviate this issue, citizen science projects have been coupled with machine learning techniques in pursuit of a more robust automated classification. However, existing approaches have neglected the fact that, apart from the data labelled by amateurs, (limited) expert knowledge of the problem is also available along with vast amounts of unlabelled data that have not yet been exploited within a unified learning framework. This paper presents an innovative learning methodology for citizen science capable of taking advantage of expert- and amateur-labelled data, featuring a transfer of labels between experts and amateurs. The proposed approach first learns from unlabelled data with a convolutional autoencoder and then exploits amateur and expert labels via the pre-training and fine-tuning of a convolutional neural network, respectively. We focus on the classification of galaxy images from the Galaxy Zoo project, from which we test binary, multi-class, and imbalanced classification scenarios. The results demonstrate that our solution is able to improve classification performance compared to a set of baseline approaches, deploying a promising methodology for learning from different confidence levels in data labelling.