Astronomers found over 5000 planets orbiting stars in our galaxy using space telescopes. The European Space Agency's Ariel telescope will observe the atmospheres of one-fifth of these exoplanets. Due to the large number of planets and complexity of the data, the mission scientists seek help from the AI and machine learning community to interpret the data.
Ariel Data Challenge
Ariel will study the light from each exoplanet’s host star after it has travelled through the planet’s atmosphere in what is known as a spectrum. The information from these spectra can help scientists investigate the chemical makeup of the planet’s atmosphere and discover more about these planets and how they formed.
Scientists involved in the Ariel mission need a new method to interpret these data. Advanced machine learning techniques could help them to understand the impact of different atmospheric phenomena on the observed spectrum. Thus: Are there any AI experts out there to join the Ariel Data Challenge and the hunt on exoplanets?
"Atmospheres as window into the formation, evolution and habitability of planets"
“Exoplanet atmospheres are a window into the formation, evolution and habitability of planets outside the Solar system - and machine learning tools provide an unexplored opportunity to study such atmospheres at a population level", says Sudeshna Boro Saikia, Tenure Track Professor for Exoplanetary Atmospheres. "Every year the Ariel Data Challenge team organizes a machine learning challenge around a key open question in the field, and provides a platform for both astronomers and machine learning experts to exchange ideas and provide their own solutions", she explains. This year the challenge is on inverse modelling https://www.ariel-datachallenge.space/ and the winners will be invited to the ECML-PKDD conference to present their results.
Key Information about the Challenge
- Start: 14 April 2023
- End: 18 June 2023
- Prize: Sponsored Ticket to ECML-PKDD or Cash Prize for top 3 winners + Invited Talk at the Ariel consortium meeting and other research institutes + Conference Proceedings for top 3 winners.
- High Performance Computing Resources: Free GPU computing resources for participants! (sponsored by DiRAC)
- Interested? Join here: www.ariel-datachallenge.space/
Partners
Ariel is European Space Agency’s upcoming exoplanet space mission and as part of the Ariel’s consortium’s machine learning working group a Machine learning Data challenge is organized every year (with a wide interdisciplinary reach). This year the University of Vienna is a partner organization; further partners are: The Alan Turing Institute, French Alternative Energies and Atomic Energy Commission, INAF - National Institute for Astrophysics, ML Analytics, UK Space Agency, STFC RAL Space, SpaceFlux, Sapienza University of Rome, University of Padova, University of Vienna, UCL DISI, Brandeis University, National Astronomical Observatory of Japan
Centre National d’Etudes Spatiales (CNES), European Research Council, UKRI Science and Technology Funding Council (STFC) Scientific Machine Learning (SciML) Division, European Space Agency and Europlanet Society support the competition.
