Our Milky Way is the only galaxy in the universe we can study in detail on a star-by-star basis, thus essential to constraint theory of galaxy formation in general. Stars serve as fossil records which we can reconstruct the Galaxy evolution and formation history with stellar ages, chemical abundances and orbits. There are multiple large-scale surveys like SDSS and Gaia making observations on stars within our Galaxy, producing big publicly available datasets. Deep learning refers to the use of artificial neural networks inspired by human brain, often doing machine learning tasks traditional algorithm cannot. In this talk, I will discuss how we use deep learning to accurately measure properties of stars from these big datasets, and to explore our own Galaxy with the measurements done by our neural network models.
Henry Leung is a fourth year PhD candidate at the Department of Astronomy & Astrophysics at the University of Toronto. His research focuses on applying deep learning methods to further the understanding of our Milky Way Galaxy. Henry previously completed his bachelor’s degree at the University of Toronto in astronomy and astrophysics. In his free time, you can find him wishing for clear skies to do some amateur astrophotography, hiking and camping.