11 October 2022
Machine learning has revolutionised data analysis in many fields of science and industry, and examples of machine learning being used in planetary and exoplanetary research are currently increasing at an almost exponential rate. Running data challenges has proved an effective way of harnessing expertise and bringing the planetary community and other data analysis fields together. In this webinar, two case studies of machine learning challenges (both currently open to participants) are presented, followed by a discussion on the benefits and difficulties of organising data challenges related to planetary science.
Ingo Waldmann – The Ariel Machine Learning Challenges. Ingo will share insights into running three interdisciplinary machine learning challenges (2020-2022) linked to deciphering exoplanets’ atmospheres, and will he discuss some of the lessons learned so far.
Nick Cox, Giacomo Nodjumi and Dan Le Corre – the EXPLORE Lunar Data Challenges. Giacomo, Nick and Dan will talk about how they went about launching a new data challenge in 2022 to identify features on the Moon and plot a lunar traverse, and will discuss some of the issues they encountered (e.g. in preparing a suitable dataset).
The use of machine and deep learning is prevalent in many fields of science and industry and is now becoming more widespread in extrasolar planet and solar system sciences. Deep learning holds many potential advantages when it comes to modelling highly non-linear data, as well as speed improvements when compared to traditional analysis and modelling techniques.
As part of the ESA Ariel Space mission, the European Conference on Machine Learning (ECML-PKDD) and NeurIPS, we have organised three very successful machine learning challenges in 2019, 2021 and 2022.
The aim was to provide new solutions to traditionally intractable problems and to foster closer collaboration between the exoplanet and machine learning communities. Often interdisciplinary approaches to long-standing problems are thwarted by jargon and a lack of familiarity. Data challenges are an excellent way to break down existing barriers and establish new links and collaborations.
Find out more: https://www.ariel-datachallenge.space
The EXPLORE Lunar Data Challenges use machine learning techniques to identify geologic features of interest on the Moon. Through the Data Challenges, the EXPLORE project aims to support the automation of planetary mapping and the identification of hazards or resources for future missions.
The Machine Learning Lunar Data Challenge is open to students, professionals and enthusiasts interested in planetary science and data processing. The EXPLORE Lunar Classroom Challenge is aimed at school students aged 10-14 year olds. The EXPLORE Public Data Challenge is open to everyone that wants to have a go.
Find out more: https://exploredatachallenges.space