In the last few years machine learning techniques have proven capable of forecasting space weather events with a much higher accuracy with respect to long-used traditional empirical and physics-based models. Even though very few operational models are currently empowered by machine learning, it appears to be unavoidable that the community will embrace in the near future such powerful techniques. Indeed, it is hard to imagine the future of space weather without machine learning. Presently, we are moving one step further from the initial ‘early-adopter’ stage, where proof-of-principles models were elaborated and tested, and more consideration is being given to the issues of reliability, uncertainty, and trustworthiness of machine learning models, finding the right balance between physics priors and data-driven discovery. In this talk the speaker presents the state-of-the-art of machine learning applications for space weather problems and discusses a few challenges and opportunities that this field presents to us.
Enrico Camporeale graduated in space plasma physics from the Queen Mary University of London. He has worked at the Los Alamos National Laboratory and the Dutch National Center for Mathematics and Computer Science. He is currently a research associate with the University of Colorado Boulder and affiliated with the NOAA Space Weather Prediction Center, in Boulder, Colorado. His research activities focus on the use of machine learning and artificial intelligence to improve the forecasting capabilities of space weather models and on data-driven discovery of space physics. Enrico Camporeale is currently an associate editor for the Journal of Space Weather and Space climate and the president-elect of the Nonlinear Geophysics section of AGU.