HAO Colloquium - Carlos José Díaz Baso, Stockholm University

Deep Learning in Solar Physics

During the last decade, deep learning has emerged as a powerful tool to extract relevant information from observations. Using very deep and complex neural networks one can improve the performance of specific tasks, sometimes much better than the conventional algorithms. In this talk, I will present some examples of how we have successfully applied deep learning to several problems in Solar Physics and highlight some results related to fast image reconstruction, 3D inversion of the Stokes parameters, and noise reduction in observational data.

Date and time: 
Wednesday, October 7, 2020 - 1:00pm to 2:00pm
Building: 
Virtual