Learning Minimal-Deviation Corrections for Multi-Dimensional Mismodelling in HEP Simulations

2026-05-08

AI Summary

In high-energy physics simulations, when only one-dimensional observational data are available, conventional methods struggle to correct high-dimensional modeling biases without distorting the intrinsic correlations among variables. This work proposes a neural network–based minimal-deviation correction approach that, under the constraint of matching only the one-dimensional target distribution, learns an optimal transformation of simulated events to align with observed data while maximally preserving the original multidimensional correlation structure. By incorporating a tailored regularization mechanism, the method achieves precise high-dimensional bias correction under limited information, balancing distributional fidelity with structural integrity. Experimental results demonstrate that the proposed approach significantly improves agreement between simulated and target distributions while effectively maintaining the underlying dependencies in the high-dimensional feature space.

Abstract

Accurate Monte Carlo (MC) modelling in high-energy physics is challenging, particularly in complex scenarios where simulations fail to reproduce observed data. In practice, experimental information is often limited to one-dimensional (1D) distributions, while mismodelling arises in a multidimensional feature space. This restricts traditional correction methods, as one-dimensional reweighting ignores correlations and fully multidimensional approaches require large target datasets. We propose a neural network-based method that operates under these constraints by learning a transformation of simulated events that reproduces the available 1D target distributions while remaining close to the original simulation. This minimal-deviation principle preserves the global correlation structure of the baseline model while enabling targeted corrections of mismodelled features. Using controlled studies with simulated pseudo-data, we show that the method improves agreement with target distributions and maintains a consistent multidimensional structure. The approach is designed for complex, high-dimensional analyses where traditional techniques are insufficient, providing a scalable way to enhance MC modelling under limited information.

Problem

• mismodelling
• Monte Carlo simulation
• high-energy physics
• multidimensional correction
• limited data

Innovation

• minimal-deviation correction
• multidimensional mismodelling
• neural network reweighting
• Monte Carlo simulation
• high-energy physics