Keele Research Repository

Context. Originating from several sources (Big Bang, stars, cosmic rays) and being strongly depleted during stellar lifetime, the lithium element is of great interest as its chemical evolution in the Milky Way is not yet well understood. To help constrain stellar and galactic chemical evolution models, numerous and precise lithium abundances are necessary for a large range of evolutionary stages, metallicities, and Galactic volume. Aims. In the age of industrial parametrization, spectroscopic surveys such as APOGEE, GALAH, RAVE, and LAMOST have used data-driven methods to rapidly and precisely infer stellar labels (atmospheric parameters and abundances). To prepare grounds for future spectroscopic surveys like 4MOST and WEAVE, we aim to apply machine–learning techniques for lithium study/measurement. Methods. We train a Convolution Neural-Network (CNN) coupling Gaia-ESO Survey iDR6 stellar labels (Teff, log(g), [Fe/H] and A(Li)) and GIRAFFE HR15N spectra, to infer the atmospheric parameters and lithium abundances for ∼ 40 000 stars. Results. We show that the CNN properly learns the physics of the stellar labels, from relevant spectral features, over a large range of evolutionary stages and stellar parameters. The lithium feature at 6707.8 Å is successfully singled out by our CNN, among the thousands of lines in the GIRAFFE HR15N setup. Rare objects like lithium-rich giants are found in our sample. Such performances are achieved thanks to a meticulously built high-quality and homogeneous training sample. Conclusions. The CNN approach is very well adapted for the next generations of spectroscopic surveys aiming at studying (among other elements) lithium, such as the 4MIDABLE-LR/HR (4MOST Milky Way disk and bulge low- and highresolution) surveys. In this context, the caveats of the machine–learning applications should be properly investigated along with realistic label uncertainties and upper limits for abundances.