Deep neural networks for analysis of space weather
keywords DEEP LEARNING, VIDEO ANALYSIS
Reference persons ENRICO MAGLI
Thesis type RESEARCH
Description This thesis addresses the problem of prediction and early detection of physical phenomena on the solar surface. Eruptions on the solar surface may give rise to coronal mass ejections, leading to significant magnetic fields that may disrupt the operation of electronic devices on the Earth or in space, including satellites. The activity has the objective of employing artificial intelligence, and in particular deep neural networks, in order to design a system able to predict and detect coronal mass ejections. The detection can be performed using images of the solar corona, as well as measurements from in-situ magnetic sensors. It is foreseen that the activity will be carried out in collaboration with the Astronomic Observatory in Pino Torinese, in order to exploit the large amount of data logged over several years, as well as ground truth data related to observed coronal mass ejections.
Required skills Candidate students should have a background in ICT or mathematics. Knowledge of neural networks (including TensorFlow environment and Python programming) is not a prerequisite, although it would help in the initial stage of the activity.
Notes We expect to use the TensorFlow environment to set up the prediction/detection system.
Deadline 18/10/2018 PROPONI LA TUA CANDIDATURA