Pan-chromatic photometric classification of supernovae from multiple surveys and transfer learning for future surveys

TL;DR

This paper introduces a method combining Gaussian processes and transfer learning to improve supernova classification across multiple surveys, addressing the challenge of limited training data in upcoming large-scale surveys.

Contribution

It presents a novel approach using Gaussian processes for uniform light curve representation and demonstrates transfer learning's effectiveness in enhancing classification accuracy for future surveys.

Findings

Achieved an AUC score of 0.859 for supernova type classification.

Transfer learning improved accuracy by up to 18% for under-represented classes.

Including photometric redshifts increased classification AUC to 0.945.

Abstract

Time-domain astronomy is entering a new era as wide-field surveys with higher cadences allow for more discoveries than ever before. The field has seen an increased use of machine learning and deep learning for automated classification of transients into established taxonomies. Training such classifiers requires a large enough and representative training set, which is not guaranteed for new future surveys such as the Vera Rubin Observatory, especially at the beginning of operations. We present the use of Gaussian processes to create a uniform representation of supernova light curves from multiple surveys, obtained through the Open Supernova Catalog for supervised classification with convolutional neural networks. We also investigate the use of transfer learning to classify light curves from the Photometric LSST Astronomical Time Series Classification Challenge (PLAsTiCC) dataset. Using convolutional neural networks to classify the Gaussian process generated representation of supernova light curves from multiple surveys, we achieve an AUC score of 0.859 for classification into Type Ia, Ibc, and II. We find that transfer learning improves the classification accuracy for the most under-represented classes by up to 18% when classifying PLAsTiCC light curves, and is able to achieve an AUC score of 0.945 when including photometric redshifts for classification into six classes (Ia, Iax, Ia-91bg, Ibc, II, SLSN-I). We also investigate the usefulness of transfer learning when there is a limited labelled training set to see how this approach can be used for training classifiers in future surveys at the beginning of operations.