CaloFlow: Fast and Accurate Generation of Calorimeter Showers with Normalizing Flows

TL;DR

CaloFlow is a novel normalizing flow-based framework that rapidly generates high-fidelity calorimeter shower simulations, outperforming GANs and VAEs in realism and training stability, offering a promising alternative to traditional methods like GEANT4.

Contribution

This paper introduces CaloFlow, the first application of normalizing flows to calorimeter shower simulation, demonstrating high accuracy, stability, and advantages over existing generative models.

Findings

CaloFlow generates calorimeter showers with high fidelity.

Classifiers struggle to distinguish CaloFlow images from real data.

CaloFlow outperforms GANs and VAEs in simulation quality.

Abstract

We introduce CaloFlow, a fast detector simulation framework based on normalizing flows. For the first time, we demonstrate that normalizing flows can reproduce many-channel calorimeter showers with extremely high fidelity, providing a fresh alternative to computationally expensive GEANT4 simulations, as well as other state-of-the-art fast simulation frameworks based on GANs and VAEs. Besides the usual histograms of physical features and images of calorimeter showers, we introduce a new metric for judging the quality of generative modeling: the performance of a classifier trained to differentiate real from generated images. We show that GAN-generated images can be identified by the classifier with nearly 100% accuracy, while images generated from CaloFlow are better able to fool the classifier. More broadly, normalizing flows offer several advantages compared to other state-of-the-art approaches (GANs and VAEs), including: tractable likelihoods; stable and convergent training; and principled model selection. Normalizing flows also provide a bijective mapping between data and the latent space, which could have other applications beyond simulation, for example, to detector unfolding.