Motivated by the high computational costs of classical simulations, machine-learned generative models can be extremely useful in particle physics and elsewhere. They become especially attractive when surrogate models can efficiently learn the underlying distribution, such that a generated sample outperforms a training sample of limited size. This kind of GANplification has been observed for simple Gaussian models. We show the same effect for a physics simulation, specifically photon showers in an electromagnetic calorimeter.
@article{arxiv.2202.07352,
title = {Calomplification -- The Power of Generative Calorimeter Models},
author = {Sebastian Bieringer and Anja Butter and Sascha Diefenbacher and Engin Eren and Frank Gaede and Daniel Hundhausen and Gregor Kasieczka and Benjamin Nachman and Tilman Plehn and Mathias Trabs},
journal= {arXiv preprint arXiv:2202.07352},
year = {2023}
}