Closing the stellar labels gap: An unsupervised, generative model for $\textit{Gaia}$ BP/RP spectra

TL;DR

This paper introduces an unsupervised deep generative model that can accurately reproduce Gaia BP/RP stellar spectra without requiring stellar labels, addressing the challenge of the massive, low-resolution spectral dataset.

Contribution

The authors develop a novel scatter variational auto-encoder capable of modeling BP/RP spectra and intrinsic scatter without labeled data, enabling analysis of large-scale stellar spectra.

Findings

Model accurately reproduces spectra where supervised methods fail

Unsupervised approach handles massive Gaia dataset effectively

Generates spectra directly from data without labels

Abstract

The recent release of 220+ million BP/RP spectra in $\textit{Gaia}$ DR3 presents an opportunity to apply deep learning models to an unprecedented number of stellar spectra, at extremely low-resolution. The BP/RP dataset is so massive that no previous spectroscopic survey can provide enough stellar labels to cover the BP/RP parameter space. We present an unsupervised, deep, generative model for BP/RP spectra: a $\textit{scatter}$ variational auto-encoder. We design a non-traditional variational auto-encoder which is capable of modeling both $(i)$ BP/RP coefficients and $(ii)$ intrinsic scatter. Our model learns a latent space from which to generate BP/RP spectra (scatter) directly from the data itself without requiring any stellar labels. We demonstrate that our model accurately reproduces BP/RP spectra in regions of parameter space where supervised learning fails or cannot be implemented.