ROSE (Red-giant Oscillations Spectra Estimator): A Modular Machine-Learning Framework for Automated Asteroseismic Characterisation. I. νmax and Δν from TESS
Abstract
Space-based photometry from Kepler and TESS has delivered oscillation spectra for hundreds of thousands of red giants, and PLATO and the Roman Galactic Bulge Time-Domain Survey will add more. We present ROSE, a modular machine-learning framework for automated asteroseismic characterisation of red giants, and apply two of its modules to the frequency of maximum oscillation power, νmax, and the large frequency separation, Δν. Each module is a neural network bundled with its own preprocessing transform and output binning, so that each parameter is measured in the representation that displays it most effectively: νmax from the power-density spectrum, and Δν from the power spectrum of the power spectrum taken over a window centred on νmax. Both are trained only on synthetic spectra generated at a one-year baseline and both return probability distributions rather than point estimates. This enables us to retrieve asymmetric uncertainty and a reliability index. Validating against held-out synthetic spectra, Kepler spectra at the training resolution and TESS continuous-viewing-zone giants, we obtain robust dispersions of 1.4, 3.3 and 4.4% in νmax and 0.35, 0.61 and 1.3% in Δν, with the νmax uncertainties close to their nominal coverage. The fully automated pipeline was applied to 274,721 TESS red-giant candidates observed for at least ten sectors, the framework returning reliable νmax for 56,224 stars and reliable Δν for 39,939, taking 8 ms per star. The sample follows Δν= 0.278 νmax0.757 with a scatter of 6.2%, and the red-clump phase stands out as a distinct overdensity, although neither module is provided an evolutionary label or asked to produce one.
Related
- Improving exoplanet mass characterisation with Bayesian model selection using the Learned Harmonic Mean Estimator
- Asteroseismic properties of young pulsators in the Perseus molecular cloud. Discovery of period-spacing patterns in HD 281229
- Asteroseismic analysis of the merger product red giant in the γ Persei system
- A Framework for Automated Multi-Source Satellite Data Analytics and LLM-Based Report Generation
- Tight asteroseismic upper limits on hot subdwarf magnetism
BibTeX
@misc{ghanghas2026rose,
title = {{ROSE (Red-giant Oscillations Spectra Estimator): A Modular Machine-Learning Framework for Automated Asteroseismic Characterisation. I. $ν_\max$ and $Δν$ from TESS}},
author = {Nipun Ghanghas and Dinil B. Palakkatharappil and Rafael A. Garcia and Shravan Hanasoge},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.10312},
eprint = {2610.10312},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.10312},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - ROSE (Red-giant Oscillations Spectra Estimator): A Modular Machine-Learning Framework for Automated Asteroseismic Characterisation. I. ν_max and Δν from TESS
AU - Nipun Ghanghas
AU - Dinil B. Palakkatharappil
AU - Rafael A. Garcia
AU - Shravan Hanasoge
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2610.10312
UR - https://arxiv.org/abs/2610.10312
AB - Space-based photometry from Kepler and TESS has delivered oscillation spectra for hundreds of thousands of red giants, and PLATO and the Roman Galactic Bulge Time-Domain Survey will add more. We present ROSE, a modular machine-learning framework for automated asteroseismic characterisation of red giants, and apply two of its modules to the frequency of maximum oscillation power, ν_max, and the large frequency separation, Δν. Each module is a neural network bundled with its own preprocessing transform and output binning, so that each parameter is measured in the representation that displays it most effectively: ν_max from the power-density spectrum, and Δν from the power spectrum of the power spectrum taken over a window centred on ν_max. Both are trained only on synthetic spectra generated at a one-year baseline and both return probability distributions rather than point estimates. This enables us to retrieve asymmetric uncertainty and a reliability index. Validating against held-out synthetic spectra, Kepler spectra at the training resolution and TESS continuous-viewing-zone giants, we obtain robust dispersions of 1.4, 3.3 and 4.4% in ν_max and 0.35, 0.61 and 1.3% in Δν, with the ν_max uncertainties close to their nominal coverage. The fully automated pipeline was applied to 274,721 TESS red-giant candidates observed for at least ten sectors, the framework returning reliable ν_max for 56,224 stars and reliable Δν for 39,939, taking 8 ms per star. The sample follows Δν= 0.278 ν_max^0.757 with a scatter of 6.2%, and the red-clump phase stands out as a distinct overdensity, although neither module is provided an evolutionary label or asked to produce one.
ER -