A New Hybrid Machine Learning Method for Stellar Parameter Inference

The advent of machine learning (ML) is revolutionary to numerous scientific disciplines, with a growing number of examples in astronomical spectroscopic inference, as ML is more powerful than traditional techniques. Here we introduce a hybrid ML (HML) method combining automatic differentiation, inte...

Full description

Saved in:
Bibliographic Details
Main Authors: Sujay Shankar, Michael A. Gully-Santiago, Caroline V. Morley
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:The Astrophysical Journal
Subjects:
Online Access:https://doi.org/10.3847/1538-4357/adcb47
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The advent of machine learning (ML) is revolutionary to numerous scientific disciplines, with a growing number of examples in astronomical spectroscopic inference, as ML is more powerful than traditional techniques. Here we introduce a hybrid ML (HML) method combining automatic differentiation, interpolation, and Bayesian optimization to infer stellar parameters of stellar spectra. We study T _eff , ${\mathrm{log}}\,(g)$ , and [Fe/H], but this method could be extended to other parameters such as [ α /Fe] (alpha element abundance), C/O (carbon–oxygen ratio), and f _sed (sedimentation efficiency). We first use blase to semiempirically recast spectra into sets of Voigt profiles. blase is run on 1314 synthetic spectra from a rectilinear subset of the PHOENIX model grid ( T _eff : [2300, 12,000] K, ${\mathrm{log}}\,(g)$ : [2, 6], [Fe/H]: [–0.5, 0], λ : [8038, 12849] Å). For 128,723 detected features, we map stellar parameters to spectral line parameters using linear interpolation. This creates the PHOENIX generator, enabling parallelized spectral synthesis. Gaussian process minimization is used to infer stellar parameters by minimizing a rms loss function. Testing 210 noise-free models ( T _eff : [3000, 11,000] K, ${\mathrm{log}}\,(g)$ : [2, 6], [Fe/H]: [−0.5, 0]), we find inference errors: T _eff : 93 K, ${\mathrm{log}}\,(g)$ : 0.24, and [Fe/H]: 0.056 for T _eff  <  7000 K, and T _eff : 347 K, ${\mathrm{log}}\,(g)$ : 0.26, and [Fe/H]: 0.16 for T _eff ≥ 7000 K. We also upload online an archive of blase models of the PHOENIX subset. This proof-of-concept study shows that semiempirical HML is a viable alternative to traditional approaches in spectroscopic inference.
ISSN:1538-4357