Bayesian classification of astronomical spectra with class uncertainties
Abstract
Context: We developed a probabilistic machine learning method with the aim of performing the O(10)-way classification of low- and high-resolution spectra of stellar and extragalactic targets for the upcoming 4MOST survey. In fulfilment of the survey requirements, this method should be able to express uncertainty in the input data as well as uncertainty introduced in its prediction. Aims: Four different methods are explored: (1) convolutional neural networks (CNNs), (2) the Dirichlet distribution, (3) Monte Carlo dropout (MCD), (4) Bayesian neural Networks (BNNs) + variational inference (VI). Training and validation was performed using labelled spectra from the SDSS database and a custom 4MOST mock dataset. All the methods were compared in terms of the same metrics: accuracy, area under the curve (AUC), expected calibration error (ECE), Shannon entropy, negative log-likelihood (NLL), Brier score, training time, and inference time. Methods: A CNN with simple architecture and about 20,000 parameters was trained to achieve classification accuracies of 91.5% on SDSS data and 92.8% on 4MOST mock data. The direct Dirichlet prediction and VI models tested provide uncertainties on class membership probabilities, but they confuse classes more often. The MCD on a CNN is found to be the most suitable; it boosts the point-estimate accuracies to 92.6% and 93.9%, while still providing fast training and sufficiently fast inference. Compared to a standard CNN, the method additionally provides well-calibrated uncertainties at marginal extra cost.
Versions
Related
- An adaptive parameter optimization method for astronomical image alignment using Bayesian optimization. I. A hierarchical search strategy for FWHM and SNR
- Feature-Based and Deep Learning Methods for Astronomical Light Curve Classification
- Contrastive Astromer: Two-Stage Contrastive Representation Learning for Astronomical Light Curve Classification
- Multi-Method Machine Learning Classification of Astronomical Sources with Rubin Data Previews: a Spectroscopy-Free Approach with DP1 and DP2
- Do little red dots really form a distinct class of astronomical objects?
BibTeX
@misc{barton2026bayesian,
title = {{Bayesian classification of astronomical spectra with class uncertainties}},
author = {Simon Barton and Martin Sahlén and Andreas Korn and Christian Glaser},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.21694},
eprint = {2609.21694},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.21694},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Bayesian classification of astronomical spectra with class uncertainties
AU - Simon Barton
AU - Martin Sahlén
AU - Andreas Korn
AU - Christian Glaser
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.21694
UR - https://arxiv.org/abs/2609.21694
AB - Context: We developed a probabilistic machine learning method with the aim of performing the O(10)-way classification of low- and high-resolution spectra of stellar and extragalactic targets for the upcoming 4MOST survey. In fulfilment of the survey requirements, this method should be able to express uncertainty in the input data as well as uncertainty introduced in its prediction. Aims: Four different methods are explored: (1) convolutional neural networks (CNNs), (2) the Dirichlet distribution, (3) Monte Carlo dropout (MCD), (4) Bayesian neural Networks (BNNs) + variational inference (VI). Training and validation was performed using labelled spectra from the SDSS database and a custom 4MOST mock dataset. All the methods were compared in terms of the same metrics: accuracy, area under the curve (AUC), expected calibration error (ECE), Shannon entropy, negative log-likelihood (NLL), Brier score, training time, and inference time. Methods: A CNN with simple architecture and about 20,000 parameters was trained to achieve classification accuracies of 91.5% on SDSS data and 92.8% on 4MOST mock data. The direct Dirichlet prediction and VI models tested provide uncertainties on class membership probabilities, but they confuse classes more often. The MCD on a CNN is found to be the most suitable; it boosts the point-estimate accuracies to 92.6% and 93.9%, while still providing fast training and sufficiently fast inference. Compared to a standard CNN, the method additionally provides well-calibrated uncertainties at marginal extra cost.
ER -