Ongoing in Geneva is the joint meeting of the European Planetary Science Congress and the Division for Planetary Sciences of the American Astronomical Society. We can abbreviate the whole thing as EPSC-DPS 2019, and you can read more about it here. We’ll track several stories here as they develop, but I notice that the European Space Agency’s ARIEL mission, which is slated to make the first large-scale survey of exoplanet atmospheres, has been supporting a Data Challenge involving removing noise from exoplanet observations. So let’s start there.
The slant here is training computers to filter out errors in collecting exoplanet data caused by starspots and by instrumentation, with two winners, James Dawson (Team SpaceMeerkat), and Vadim Borisov (Team major_tom), announced yesterday in Geneva. All told, 112 teams registered for the competition, a heartening number illustrative of the growing interest in computational statistics and machine learning among exoplanet researchers. The top five teams in this first international Machine Learning Data Challenge will present their methodologies to the European Conference on Machine Learning (ECML-PKDD 2019) on Friday.
Says Nikos Nikolaou (UCL Centre for Exochemistry Data), who created the competition:
“The outcomes of the competition exceeded our expectations, both in terms of ...