2023A&A...675A..68V

Context. Membership studies characterising open clusters (OCs) with Gaia data - most of them using Gaia Data Release 2 (DR2) - have so far been limited at the faint end to magnitude G = 18 due to astrometric uncertainties.
Aims. Our goal is to extend current OC membership lists with faint members and to characterise the low-mass end. These low-mass members are important for many applications, in particular for ground-based spectroscopic surveys.
Methods. We use a deep neural network architecture to learn the distribution of highly reliable OC member stars around known clusters. We then use the trained network to estimate new OC members based on their similarities in a high dimensional space, their five-dimensional astrometry, and information from the three photometric bands.
Results. Due to the improved astrometric precision of Gaia Data Release 3 (DR3) with respect to DR2, we are able to homogeneously detect new faint member stars (G > 18) for the known OC population.
Conclusions. Our methodology can provide extended membership lists for OCs down to the limiting magnitude of Gaia, which will enable further studies to characterise the OC population; such as estimation of their masses and dynamics. These extended membership lists are also ideal target lists for forthcoming ground-based spectroscopic surveys.