Machine learning applied to image interiors of Red-giant stars

One of the fundamental problems in astrophysics is to understand how stars evolve over their lifetimes. The transition to the red-giant phase is complex, involving dramatic changes to the interior structure and rotation, and is very poorly understood. The Kepler mission (2009) has provided an enormous trove of seismic data - of thousands of red giants giving us an opportunity to understand this transition. I will be discussing the Asymptotic theory of Oscillations that models the asteroseimsic spectra. I will present an efficient machine learning algorithm that we developed to analyse this seismic data. I will present the algorithm's results on the Kepler data and compare those results with the published results.