Konstantin Matchev - A Machine Learning Journey in a World of Alien Planets

Wednesday 28 May 2025, 15:30 → 17:00 Europe/Rome

via della Vasca Navale 84

 I will review the use of transit spectroscopy for the characterization of the chemical composition and physical parameters of the atmospheres of exoplanets. I will describe some of the machine learning techniques, which allowed our team to win the Ariel Machine Learning Data Challenge at the 2022 NeurIPS conference. I will also review our recent work on the application of machine learning for detection of anomalous transit spectra, with the goal of identifying planets with unusual chemical composition and even searching for unknown biosignatures.