Deep learning exoplanets and the solar system

jeudi 7 février 2019, par Ingo P. Waldmann (Deputy Director UCL Centre of Space Exoplanet Data, Dept. of Physics & Astronomy, University College London)

Lundi 4 mars 2019 à 14h00 , Lieu : Amphithéâtre Evry Schatzman, bâtiment 18

The field of exoplanetary spectroscopy is as fast moving as it is new. Analysing currently available observations of exoplanetary atmospheres often invoke large and correlated parameter spaces that can be difficult to map or constrain. This is true for both : the data analysis of observations as well as the theoretical modelling of their atmospheres. Issues of low signal-to-noise data and large, non-linear parameter spaces are nothing new and commonly found in many fields of engineering and the physical sciences. Recent years have seen vast improvements in statistical data analysis and machine learning that have revolutionised fields as diverse as telecommunication, pattern recognition, medical physics and cosmology. In many aspects, data mining and non-linearity challenges encountered in other data intensive fields are directly transferable to the field of extrasolar planets as well as planetary sciences. In this seminar, I will discuss our new deep learning framework, ExoGAN (Tzingales & Waldmann, 2018, AJ), designed to address some of these atmospheric modelling challenges using generative adversarial networks. I will then proceed to discuss our new hyper-spectral image classification code, PlanetNET (Waldmann & Griffith, in press, Nat. Astr.), able to automatically and accurately map Saturn’s clouds using Cassini/VIMS data. As we firmly move into the era of ‘big data’ for both planetary (e.g. Juno) and exoplanetary sciences (e.g. JWST, Ariel), intelligent algorithms will play an important part in facilitating the analysis of these rich data sets in the future.

0 | 10 | 20