Parameters for > 300 million Gaia stars: Bayesian inference vs. machine learning

TL;DR

This paper compares Bayesian inference and machine learning methods, particularly XGBoost, for estimating stellar parameters from Gaia DR3 data, demonstrating that machine learning can produce competitive results efficiently for over 300 million stars.

Contribution

It introduces a machine learning approach using XGBoost for stellar parameter estimation from Gaia DR3, showing its effectiveness compared to traditional Bayesian methods.

Findings

Machine learning achieves competitive accuracy with Bayesian methods.

Simple neural networks and tree-based algorithms perform well on large Gaia data.

A new Gaia DR3 stellar-parameter catalogue will be released using XGBoost.

Abstract

The Gaia Data Release 3 (DR3), published in June 2022, delivers a diverse set of astrometric, photometric, and spectroscopic measurements for more than a billion stars. The wealth and complexity of the data makes traditional approaches for estimating stellar parameters for the full Gaia dataset almost prohibitive. We have explored different supervised learning methods for extracting basic stellar parameters as well as distances and line-of-sight extinctions, given spectro-photo-astrometric data (including also the new Gaia XP spectra). For training we use an enhanced high-quality dataset compiled from Gaia DR3 and ground-based spectroscopic survey data covering the whole sky and all Galactic components. We show that even with a simple neural-network architecture or tree-based algorithm (and in the absence of Gaia XP spectra), we succeed in predicting competitive results (compared to Bayesian isochrone fitting) down to faint magnitudes. We will present a new Gaia DR3 stellar-parameter catalogue obtained using the currently best-performing machine-learning algorithm for tabular data, XGBoost, in the near future.