Use the 4S (Signal-Safe Speckle Subtraction): Explainable Machine Learning Reveals the Giant Exoplanet AF Lep b in High-contrast Imaging Data from 2011

Abstract The main challenge of exoplanet high-contrast imaging (HCI) is to separate the signal of exoplanets from their host stars, which are many orders of magnitude brighter. HCI for ground-based observations is further exacerbated by speckle noise originating from perturbations in Earth’s atmosphere and imperfections in the telescope optics. Various data postprocessing techniques are used to remove this speckle noise and reveal the faint planet signal. Often, however, a significant part of the planet signal is accidentally subtracted together with the noise. In the present work, we use explainable machine learning to investigate the reason for the loss of the planet signal for one of the most used postprocessing methods: principal component analysis (PCA). We find that PCA learns the shape of the telescope point-spread function for high numbers of PCA components. This representation of the noise captures not only the speckle noise but also the characteristic shape of the planet signal. Building on these insights, we develop a new postprocessing method (4S) that constrains the noise model to minimize this signal loss. We apply our model to 11 archival HCI data sets from the Very Large Telescope NACO instrument in the L ’ band and find that our model consistently outperforms PCA. The improvement is largest at close separations to the star (≤4 λ / D ), providing up to 1.5 mag deeper contrast. This enhancement enables us to detect the exoplanet AF Lep b in data from 2011, 11 yr before its subsequent discovery. We present updated orbital parameters for this object.