Speaker
Description
In recent times, the amount of data collected from high-energy experiments has increased continuously. In order to match this amount of data, significantly more Monte Carlo data is required as well. These simulations often take a long time and occupy a lot of computing power. For complex simulation steps, such as the simulation of electromagnetic calorimeters, neural networks can offer considerable acceleration and thus make these steps not only faster, but also more resource-efficient. In this talk, use cases of neural networks in Monte Carlo simulations are presented and the development is explained using the forward endcap of the EMC of the PANDA experiment. Furthermore, possibilities are presented to use the results of complex partial wave analyses to produce Monte Carlo events more accurately and faster.
