Accelerating Monte Carlo event generation -- rejection sampling using neural network event-weight estimates

TL;DR

This paper introduces a neural network-based two-stage unweighting method that accelerates Monte Carlo event generation for complex scattering processes while maintaining unbiased sampling, achieving up to tenfold speed improvements.

Contribution

A novel neural network surrogate-based unweighting procedure that significantly speeds up Monte Carlo event generation without bias.

Findings

Achieves up to ten times faster event generation.

Validates the method on high-multiplicity LHC processes.

Maintains unbiased sampling with neural network surrogate.

Abstract

The generation of unit-weight events for complex scattering processes presents a severe challenge to modern Monte Carlo event generators. Even when using sophisticated phase-space sampling techniques adapted to the underlying transition matrix elements, the efficiency for generating unit-weight events from weighted samples can become a limiting factor in practical applications. Here we present a novel two-staged unweighting procedure that makes use of a neural-network surrogate for the full event weight. The algorithm can significantly accelerate the unweighting process, while it still guarantees unbiased sampling from the correct target distribution. We apply, validate and benchmark the new approach in high-multiplicity LHC production processes, including $Z/W$+4 jets and $t\bar{t}$+3 jets, where we find speed-up factors up to ten.