Star formation history of Galactic Star Forming Regions: preliminary results

One of the major astrophysical challenges is to determine the star formation history and the rate of star formation of Galactic Star Forming Regions (SFRs). These are the fundamental ingredients for understanding the dominant effects and the physical mechanisms that drive the star formation process. Additionally, they provide valuable insights into the structure and evolution of the youngest stellar component of the Milky Way. The main challenge consists in accurately estimating fundamental astrophysical parameters - such as effective temperatures, luminosities and stellar ages - for large samples of young stellar objects (YSO), from which the star formation history can be inferred. In the era of Gaia and other optical/NIR large photometric surveys, YSOs in SFRs can be reliably identified, but characterising them remains challenging without spectroscopic data. In this talk, we present a model based on a deep neural network that can predict the effective temperatures of YSOs in the range 3100-7000 K. This machine learning approach leverages both spectroscopic and photometric variables as input features for the training set, with data from the Gaia-ESO (GES), Gaia DR3, and 2MASS surveys. We apply this model to an all-sky catalog of low-mass YSOs to predict their effective temperatures. Using standard astrophysical techniques and novel theoretical models, we derive the stellar ages of these objects. A preliminary statistical analysis of the age distributions of the most prominent Galactic SFRs will be presented and discussed.