Improving FastCaloGAN Simulation of Pion Calorimeter Showers

FastCaloGAN uses unsupervised machine learning techniques to provide a fast simulation of the full ATLAS calorimeter system. In this project, the effects of modifying various elements of the FastCaloGAN's architecture were studied and this report describes some of the most successful modificat...

Descripción completa

Detalles Bibliográficos
Autor principal: Barber, Matthew William
Lenguaje:eng
Publicado: 2021
Materias:
Particle Physics - Experiment
Acceso en línea:http://cds.cern.ch/record/2780098
Descripción
Sumario:FastCaloGAN uses unsupervised machine learning techniques to provide a fast simulation of the full ATLAS calorimeter system. In this project, the effects of modifying various elements of the FastCaloGAN's architecture were studied and this report describes some of the most successful modifications. Overall, we were able to achieve significantly better performance with greatly reduced computational resources.

Ejemplares similares