An Artificial Neural Network for on-board event pre-processing of Gamma-Ray Burst Observations
Presenter: Christopher Weinert
Title: Estimating Galaxy Cluster Mass Accretion Rates from Observations using Machine Learning
Date/Time: Monday, June 17th, 2:30 - 4:00 PM; Thursday, June 20th, 3:30 - 5:00 PM
Abstract: The aim is to have a Machine or Deep Learning model which can classify high-energetic (MeV to GeV range) astrophysical phenomena – with a particular focus on Gamma-Ray bursts and its potential progenitors. This model is planned to be implemented in a detector for on-the-fly predictions. At the moment, I pursue a feature-based classification: to derive new features from existing data, which are more distinctive and significant for the learning process of the AI. Within this project I test the performance on a large number of different models in Machine Learning and Deep Learning. Lastly, I want to modify my selection of the best models even further with Uncertainty Estimations and Quantifications.