ML‐Helio: An Emerging Community at the Intersection Between Heliophysics and Machine Learning

Abstract

The advancements and breakthroughs achieved in the last 5–10 years in Artificial Intelligence and machine learning (ML) have not gone unnoticed in the scientific community. The body of literature that borrows techniques from ML has steadily grown in all fields of physics. Space physics is particularly well posed to exploit ML due to the large amount of (often under scrutinized) data accumulated over the last few decades. Indeed, ML techniques can offer insights into the data that might enhance our understanding of physical mechanisms. Many of the pioneering studies on the use of ML in Space Physics have been led by several individuals who have independently taken the burden of moving out of their comfort zone to climb the steep slope of learning new jargon, new methodologies, and new coding skills. Such early adopters have recently convened in Amsterdam for the first conference on machine learning in heliophysics. The conference has laid the foundation for a new emerging community, and this commentary summarizes the discussions and steps taken to make such community flourish. Discussion around a new community focused on machine learning in heliophysics