Kepler Star Classification with Convolutional Neural Network on Light Curve Frequency Power Spectra.

Deep learning classification in asteroseismology
Automated methods do exist, however considerable effort is required into defining and acquiring features such as the observed period spacing ΔP (Bedding et al. 2011; Stello et al. 2013), the asymptotic period spacing ΔΠ1 (Bedding et al. 2011; Mosser et al. 2014; Vrard, Mosser & Samadi 2016) or the structure of mixed modes (Elsworth et al. 2016) in order to separate the populations. Furthermore, these methods require relatively high signal-to-noise data.