A Comprehensive Evaluation of Generative Models in Calorimeter Shower Simulation

TL;DR

This paper rigorously evaluates three deep generative models for particle shower simulation in calorimeters, highlighting their strengths and shortcomings in replicating detailed physics data, and exploring inference precision impacts.

Contribution

It provides a comprehensive, standardized assessment of generative models for calorimeter simulation, comparing their performance and analyzing inference precision effects.

Findings

CaloDiffusion and CaloScore outperform others in accuracy

Models still have significant gaps compared to Geant4 data

Full vs mixed precision inference impacts model performance

Abstract

The pursuit of understanding fundamental particle interactions has reached unparalleled precision levels. Particle physics detectors play a crucial role in generating low-level object signatures that encode collision physics. However, simulating these particle collisions is a demanding task in terms of memory and computation which will be exasperated with larger data volumes, more complex detectors, and a higher pileup environment in the High-Luminosity LHC. The introduction of "Fast Simulation" has been pivotal in overcoming computational bottlenecks. The use of deep-generative models has sparked a surge of interest in surrogate modeling for detector simulations, generating particle showers that closely resemble the observed data. Nonetheless, there is a pressing need for a comprehensive evaluation of their performance using a standardized set of metrics. In this study, we conducted a rigorous evaluation of three generative models using standard datasets and a diverse set of metrics derived from physics, computer vision, and statistics. Furthermore, we explored the impact of using full versus mixed precision modes during inference. Our evaluation revealed that the CaloDiffusion and CaloScore generative models demonstrate the most accurate simulation of particle showers, yet there remains substantial room for improvement. Our findings identified areas where the evaluated models fell short in accurately replicating Geant4 data.