TY  - JOUR
PB  - SciPost Foundation
DO  - 10.21468/SciPostPhys.18.3.081
TI  - Normalizing flows for high-dimensional detector simulations
PY  - 2025/03/05
UR  - https://scipost.org/SciPostPhys.18.3.081
JF  - SciPost Physics
JA  - SciPost Phys.
VL  - 18
IS  - 3
SP  - 081
A1  - Ernst, Florian
AU  - Favaro, Luigi
AU  - Krause, Claudius
AU  - Plehn, Tilman
AU  - Shih, David
AB  - Whenever invertible generative networks are needed for LHC physics, normalizing flows show excellent performance. In this work, we investigate their performance for fast calorimeter shower simulations with increasing phase space dimension. We use fast and expressive coupling spline transformations applied to the CaloChallenge datasets. In addition to the base flow architecture we also employ a VAE to compress the dimensionality and train a generative network in the latent space. We evaluate our networks on several metrics, including high-level features, classifiers, and generation timing. Our findings demonstrate that invertible neural networks have competitive performance when compared to autoregressive flows, while being substantially faster during generation.
ER  -

@Article{10.21468/SciPostPhys.18.3.081,
	title={{Normalizing flows for high-dimensional detector simulations}},
	author={Florian Ernst and Luigi Favaro and Claudius Krause and Tilman Plehn and David Shih},
	journal={SciPost Phys.},
	volume={18},
	pages={081},
	year={2025},
	publisher={SciPost},
	doi={10.21468/SciPostPhys.18.3.081},
	url={https://scipost.org/10.21468/SciPostPhys.18.3.081},
}