EarthCube RCN Workshop

Machine Learning in Heliophysics and Space Weather Forecasting: Advances, Perspectives and Synergies

16-17 January 2020

New Jersey Institute of Technology, Newark, NJ

Workshop Organizing Committee

Gelu Nita (NJIT), Manolis Georgoulis (GSU), Alexander Kosovichev (NJIT), Haimin Wang (NJIT), Vincent Oria (NJIT), Jason Wang (NJIT), Petrus Martens (GSU), Rafal Angryk (GSU)


NJIT Guest Wireless Network

Please follow these instructions to connect to the NJIT guest wireless internet network while on campus:


The main goals of the workshop are to discuss recent critical developments and prospects of the application of machine and / or deep learning techniques for data analysis, modeling and forecasting in Heliophysics and to shape a strategy for further development within the topical Research Coordinated Network. A desired outcome of this activity is to collectively produce a white paper that will be submitted to NSF and NASA, indicating (i) the state of the art in ML / DL applications for Heliophysics and (ii) clear-cut suggestions for future actions, contributions and synergies.

Registered Participants


New Jersey Institute of Technology

Robert Treat Hotel, Newark

NJIT Campus

CAMPUS-MAP-2017_09_08.pdf

Dramatic increase in observational data flow from ground and space telescopes (e.g., SDO, IRIS, Hinode, Parker Probe, BBSO, etc.), as well as from upcoming observational facilities, such as the 4-meter solar telescope DKIST, provides unique opportunity to probe the solar dynamics from the interior to corona with high spatial and temporal resolution. At the same time it makes a great challenge to analyze the large multi-dimensional data sets, and to interpret the observed highly nonlinear processes. The modern data analysis involves detailed NLTE spectro-polarimetric diagnostics. Recent advances in 3D radiative modeling make it possible to reproduce the dynamics of subsurface magnetoconvection layers and the solar atmosphere with a high degree of realism, providing tools for physics-based interpretation of observational data. In current state of the field, cross-validation and verification of numerical models, data analysis techniques, and physical interpretation of detected features become extremely important. The goal of the working group is to create a platform for interaction of observers, data analysts and modelers to identify key points and challenges that need to be addressed for advancing the knowledge of physical processes on the Sun, which control the solar energy output and magnetic activity. The working group activities will include discussions of the current challenges, new ways to efficiently resolve them, promoting cross-disciplinary activity, establishing collaborations, writing white papers and research proposals.

Founding working group members: Irina Kitiashvili (NASA Ames/BAERI), Serena Criscuoli (NSO), Alexander Kosovichev (NJIT)