Unfolding with Generative Adversarial Networks

2018/06/01 by K. Datta, Datta, Kaustuv, D. Kar +3 · 8 citations
Computer Science · Physics and Astronomy · #Data Analysis #FOS: Physical sciences #Gaussian Processes and Bayesian Inference #High Energy Physics - Experiment (hep-ex) #High Energy Physics - Phenomenology (hep-ph) #High-Energy Particle Collisions Research #Particle physics theoretical and experimental studies #Statistics and Probability (physics.data-an)

paper · pdf · doi:10.48550/arxiv.1806.00433

openalex publication_date 2018/06/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

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.

Citations

• Improved Techniques for Training GANs

Cited by

• Machine Learning-based Unfolding for Cross Section Measurements in the Presence of Nuisance Parameters
• LHC analysis-specific datasets with Generative Adversarial Networks
• Generative Unfolding of Jets and Their Substructure
• Analysis-ready Generative Unfolding
• Response Matrix Estimation in Unfolding Differential Cross Sections
• Toward an event-level analysis of hadron structure using differential programming
• Reconstructing Sparticle masses at the LHC using Generative Machine Learning
• A Practical Guide to Unbinned Unfolding

Related

• Machine learning as an instrument for data unfolding
• A Practical Guide to Unbinned Unfolding
• Event-by-event Comparison between Machine-Learning- and Transfer-Matrix-based Unfolding Methods
• An unfolding method based on conditional Invertible Neural Networks (cINN) using iterative training
• Multidimensional Deconvolution with Profiling