Studying Stellar Spectra using Machine Learning

Stellar astronomy has entered a remarkable era, with an ensemble of spectroscopic surveys providing spectra for millions of stars. With the dramatic increase in data, there is an opportunity to use the data in concert to understand the parameters that control its variance, and its information content. I will showcase the utility of simple and interpretable models built using spectral data to (i) make precision measurements from stars, (ii) determine the information content in stellar spectra and (iii) find stars with anomalous abundances given their evolutionary state.