Milky Way open cluster age determination with spectroscopic priors and machine learning-accelerated inference

Context . Open clusters are excellent tracers of the Galactic thin disk, providing insights into its formation and evolution. Precise cluster age determination is essential for studying cluster evolution as well as the evolution of the disk as a whole. Aims . With this work, our aim was to produce an updated catalog of cluster parameters (age, extinction, distance) by constraining cluster ages via newly available spectroscopic measurements of metallicity. Methods . Using high-quality astrometry from Gaia Data Release 3, we derived new cluster memberships via a machine learning-based clustering algorithm (AstroLink). We compiled metallicity measurements from multiple ground-based spectroscopic surveys (Gaia-ESO, GALAH DR3, APOGEE-OCCAM) and Gaia GSP-Spec into a uniform scale. The ages of 320 clusters were determined using a machine learning-accelerated approximate Bayesian isochrone-fitting code (based on ASteCA), adopting a multiband approach that incorporates Gaia XP and synthetic photometry. Results . This work resulted in a robust sample of open clusters with precisely determined parameters and uncertainties. In addition, this catalog may provide a foundation for machine learning models to infer parameters for a larger sample of objects without spectroscopic data, which is to be explored in future work.