Thursday 23 May 2019: [Seminar] BRINGING SPACE EXPLORATION AI APPLICATIONS BACK TO EARTH: Machine Learning for Exoplanet research Lessons and results from NASA Frontier Development Lab
Dr Daniel Angerhausen - University of Bern, Switzerland
4th Floor, Physics Building 11:30-12:30
In my presentation I will give a short introduction to the science of extrasolar planets, in particular the technique of transit, eclipse and phasecurve spectro-photometry. I will describe my various projects in this emerging field using state of the art spectroscopic and photometric instruments on the largest ground based telescopes, the 'flying telescope' SOFIA (Stratospheric Observatory for Infrared Astronomy) and the Kepler and Hubble space telescopes. In addition I will present results from NASA's Frontier Development Lab 2018, an Artificial Intelligence/Machine Learning incubator tackling challenges in this field, such as planet search in TESS data or retrieval of exoplanet spectra. A particular focus will be on two data sets produced at FDL 2018: a set of 3 million exoplanet spectra calculated with the GSFC Planetary Spectrum Generator (PSG) and a set of 150.000 exoplanet atmospheres computed with the ATMOS code.