How to GAN Event Unweighting

TL;DR

This paper proposes an improved generative network approach to unweight event generation, aiming to significantly accelerate LHC simulations by addressing a key bottleneck in the standard unweighting procedure.

Contribution

It introduces a novel method that enhances the unweighting process in event generation using neural networks, potentially increasing simulation speed for high-energy physics.

Findings

Demonstrates potential for substantial speed-up in event simulation

Addresses a known bottleneck in standard unweighting procedures

Provides a framework for integrating neural networks into LHC simulation workflows

Abstract

Event generation with neural networks has seen significant progress recently. The big open question is still how such new methods will accelerate LHC simulations to the level required by upcoming LHC runs. We target a known bottleneck of standard simulations and show how their unweighting procedure can be improved by generative networks. This can, potentially, lead to a very significant gain in simulation speed.