StelNet: Hierarchical Neural Network for Automatic Inference in Stellar Characterization

Abstract

Characterizing the fundamental parameters of stars from observations is crucial for studying the stars themselves, their planets, and the galaxy as a whole. Stellar evolution theory predicting the properties of stars as a function of stellar age and mass enables translating observables into physical stellar parameters by fitting the observed data to synthetic isochrones. However, the complexity of overlapping evolutionary tracks often makes this task numerically challenging, and with a precision that can be highly variable, depending on the area of the parameter space the observation lies in. This work presents StelNet, a Deep Neural Network trained on stellar evolutionary tracks that quickly and accurately predicts mass and age from absolute luminosity and effective temperature for stars with close to solar metallicity. The underlying model makes no assumption on the evolutionary stage and includes the pre-main sequence phase. We use bootstrapping and train many models to quantify the uncertainty of the model. To break the model's intrinsic degeneracy resulting from overlapping evolutionary paths, we also built a hierarchical model that retrieves realistic posterior probability distributions of the stellar mass and age. We further test and train StelNet using a sample of stars with well-determined masses and ages from the literature.

arXiv comment17 pages, 15 figures, accepted to AJ

References in corpus (39)

• On the variation of the Initial Mass Function
• Modules for Experiments in Stellar Astrophysics (MESA)
• PARSEC: stellar tracks and isochrones with the PAdova and TRieste Stellar Evolution Code
• Modules for Experiments in Stellar Astrophysics (MESA): Giant Planets, Oscillations, Rotation, and Massive Stars
• Modules for Experiments in Stellar Astrophysics (MESA): Binaries, Pulsations, and Explosions
• MESA Isochrones and Stellar Tracks (MIST). I: Solar-Scaled Models
• MESA Isochrones and Stellar Tracks (MIST) 0: Methods for the construction of stellar isochrones
• Improved Age Estimation for Solar-Type Dwarfs Using Activity-Rotation Diagnostics
• The Milky Way Tomography with SDSS: I. Stellar Number Density Distribution
• The Revised TESS Input Catalog and Candidate Target List
• How to Constrain Your M Dwarf: measuring effective temperature, bolometric luminosity, mass, and radius
• Gaia Data Release 2: first stellar parameters from Apsis
• Structure and Evolution of Nearby Stars with Planets II. Physical Properties of ~1000 Cool Stars from the SPOCS Catalog
• Gaia FGK Benchmark Stars: Effective temperatures and surface gravities
• Isochrone ages for field dwarfs: method and application to the age-metallicity relation
• First radius measurements of very low mass stars with the VLTI
• Constraining fundamental stellar parameters using seismology. Application to Alpha Centauri AB
• The Ages of Early-Type Stars: Strömgren Photometric Methods Calibrated, Validated, Tested, and Applied to Hosts and Prospective Hosts of Directly Imaged Exoplanets
• Asteroseismology and calibration of alpha Cen binary system
• New distances to RAVE stars
• Analysis of alpha Centauri AB including seismic constraints
• Stellar distances from spectroscopic observations: a new technique
• Fundamental Parameters of Main-Sequence Stars in an Instant with Machine Learning
• The radii and limb darkenings of Alpha Centauri A and B - Interferometric measurements with VLTI/PIONIER
• The Alpha Centauri Binary System: Atmospheric Parameters and Element Abundances
• Quality of Uncertainty Quantification for Bayesian Neural Network Inference
• Distance determination for RAVE stars using stellar models
• Classically and Asteroseismically constrained 1D Stellar Evolution Models of $\alpha$ Centauri A and B using Empirical Mixing Length Calibrations
• MINESweeper: Spectrophotometric Modeling of Stars in the Gaia Era
• Implicit Weight Uncertainty in Neural Networks
• A Bayesian approach to the modelling of alpha Cen A
• A new seismic analysis of Alpha Centauri
• Close stellar conjunctions of alpha Centauri A and B until 2050 - An mK = 7.8 star may enter the Einstein ring of alpha Cen A in 2028
• Stellar ages, masses and radii from asteroseismic modeling are robust to systematic errors in spectroscopy
• Computation of the Memory Functions in the Generalized Langevin Models for Collective Dynamics of Macromolecules
• MESA models with magnetic braking
• Asteroseismic determination of fundamental parameters of sun-like stars using multi-layered neural networks
• Deep Evidential Regression
• Gaia Benchmark stars and their twins in the Gaia-ESO Survey