Discovering sets of pulsators with machine learning and finding automated ways to probe their physics.
The Kepler and TESS space missions have revolutionized the field of asteroseismology by delivering light curves for millions of stars. The challenge now lies in leveraging the information present in these stars and using it to improve our physical knowledge of the internal workings of stars. Machine learning proves to be the ideal solution in this case as its predictions improve with the amount of data available. We therefore developed a machine learning pipeline with both a supervised and unsupervised learning component that allows us to 1) classify the light curves according to their stellar variability type and 2) provide us with new insights into the physical relations that govern them. In this talk, we will in particular focus on how our methodology can be used to study the physical interplay between the rotation and pulsations of γ Doradus stars.