Generative models uncertainty estimation

TL;DR

This paper introduces three methods for estimating the uncertainty of generative models in high-energy physics simulations, addressing the challenge of model reliability in sparse data regions, and demonstrates their application on LHCb RICH data.

Contribution

It proposes novel uncertainty estimation techniques for generative models in physics simulations, including calibration methods, applicable both inside and outside training regions.

Findings

Effective uncertainty estimation methods demonstrated on LHCb RICH simulation.

Calibration techniques improve model reliability in sparse data regions.

Methods enhance trustworthiness of data-driven physics simulations.

Abstract

In recent years fully-parametric fast simulation methods based on generative models have been proposed for a variety of high-energy physics detectors. By their nature, the quality of data-driven models degrades in the regions of the phase space where the data are sparse. Since machine-learning models are hard to analyse from the physical principles, the commonly used testing procedures are performed in a data-driven way and can't be reliably used in such regions. In our work we propose three methods to estimate the uncertainty of generative models inside and outside of the training phase space region, along with data-driven calibration techniques. A test of the proposed methods on the LHCb RICH fast simulation is also presented.