Revealing the Milky Way's most recent major merger with a Gaia EDR3 catalogue of machine-learned line-of-sight velocities

Machine learning can play a powerful role in inferring missing line-of-sight velocities from astrometry in surveys such as Gaia. In this paper, we apply a neural network to Gaia Early Data Release 3 (EDR3) and obtain line-of-sight velocities and associated uncertainties for similar to 92 million stars. The network, which takes as input a star's parallax, angular coordinates, and proper motions, is trained and validated on similar to 6.4 million stars in Gaia with complete phase-space information. The network's uncertainty on its velocity prediction is a key aspect of its design; by properly convolving these uncertainties with the inferred velocities, we obtain accurate stellar kinematic distributions. As a first science application, we use the new network-completed catalogue to identify candidate stars that belong to the Milky Way's most recent major merger, Gaia-Sausage-Enceladus (GSE). We present the kinematic, energy, angular momentum, and spatial distributions of the similar to 450 000 GSE candidates in this sample, and also study the chemical abundances of those with cross matches to GALAH and APOGEE. The network's predictive power will only continue to improve with future Gaia data releases as the training set of stars with complete phase-space information grows. This work provides a first demonstration of how to use machine learning to exploit high-dimensional correlations on data to infer line-of-sight velocities, and offers a template for how to train, validate, and apply such a neural network when complete observational data is not available.

Automated summary

Machine learning is used to infer missing line-of-sight velocities from astrometry in surveys like Gaia. The neural network applied to Gaia EDR3 predicts line-of-sight velocities and uncertainties for around 92 million stars using their parallax, angular coordinates, and proper motions. The network's uncertainty on velocity prediction is crucial for obtaining accurate stellar kinematic distributions. This work demonstrates the use of machine learning to exploit high-dimensional correlations for inferring line-of-sight velocities and provides a template for training and applying such a neural network when complete observational data is lacking.