Systematic Evaluation of Generative Machine Learning Capability to Simulate Distributions of Observables at the Large Hadron Collider

Abstract

Monte Carlo simulations are a crucial component when analysing the Standard Model and New physics processes at the Large Hadron Collider. This paper aims to explore the performance of generative models for complementing the statistics of classical Monte Carlo simulations in the final stage of data analysis by generating additional synthetic data that follows the same kinematic distributions for a limited set of analysis-specific observables to a high precision. Several deep generative models are adapted for this task and their performance is systematically evaluated using a well-known benchmark sample containing the Higgs boson production beyond the Standard Model and the corresponding irreducible background. The paper evaluates the autoregressive models and normalizing flows and the applicability of these models using different model configurations is investigated. The best performing model is chosen for a further evaluation using a set of statistical procedures and a simplified physics analysis. By implementing and performing a series of statistical tests and evaluations we show that a machine-learning-based generative procedure can be used to generate synthetic data that matches the original samples closely enough and that it can therefore be incorporated in the final stage of a physics analysis with some given systematic uncertainty.