Victor Ksoll (Germany)
v.ksoll @ stud.uni-heidelberg.de
Characterising Pre-Main-Sequence Stars in the Large Magellanic Cloud with Machine Learning Techniques
Dedicated photometric surveys with the Hubble Space Telescope (HST) have provided unprecedented coverage of the two most active star-forming regions in the Large Magellanic Cloud. The Hubble Tarantula Treasury Project (HTTP; PI Sabbi), the panchromatic survey of the star-burst region of 30 Doradus, combined with the HST monochromatic survey to measure proper motions of massive stellar runaways, has delivered the complete young stellar population of the Tarantula nebula down to the half solar mass limit. MYSST: Measuring Young Stars in Space and Time (PI Gouliermis), the deepest dual-band HST stellar survey in the LMC, combined with the VLT-FLAMES spectroscopic survey of massive stars, delivers the richest young stellar inventory of the impressive star-forming complex N44, targeting recently born stars down to the hydrogen burning limit. We use the deep stellar catalogues of HTTP and MYSST to identify all the pre-main-sequence (PMS) stars in the respective regions. The photometric distinction of these stars from more evolved populations is, however, not a trivial task due to several factors that alter their colour-magnitude diagram positions. To overcome this hurdle, we employ Machine Learning Classification techniques, including Random Forests and Support Vector Machines (SVM), on the HTTP and MYSST surveys to unveil their PMS stellar content. Our methodology consists of 1) carefully selecting the most probable low-mass PMS stellar population of a prominent star forming cluster within the observed fields, 2) using these samples to train classification algorithms to build predictive models for PMS stars and 3) applying these models to identify the most probable PMS content across the entire observed regions. Furthermore we make use of Deep Learning techniques, developing an Invertible Neural Network (INN), in order to predict the fundamental physical parameters of age and mass of the identified young stars and evaluate the spatial variations of these parameters across the entire star-forming complexes. This outcome, in combination with the high-mass census of the regions, will provide an original understanding of how star formation proceeds in space and time.
Supervisors: Ralf Klessen (ITA)