Unfolding with Generative Adversarial Networks

TL;DR

This paper introduces a novel unfolding method using a modified Generative Adversarial Network (MSGAN) to correct detector-level data to particle-level, demonstrating comparable performance to existing techniques across diverse distributions.

Contribution

It presents a new machine learning-based unfolding technique with MSGAN, showcasing its potential as a state-of-the-art approach for modeling detector effects.

Findings

Performs comparably to current unfolding methods

Effective across various distribution shapes

Proof-of-principle demonstration of MSGAN in unfolding

Abstract

Correcting measured detector-level distributions to particle-level is essential to make data usable outside the experimental collaborations. The term unfolding is used to describe this procedure. A new method of unfolding data using a modified Generative Adversarial Network (MSGAN) is presented here. Applied to various distributions with widely different shapes, it performs roughly at par with currently used methods. This is a proof-of-principle demonstration of a state-of-the-art machine learning method that can be used to model detector effects well.