• Corpus ID: 51682314

Unfolding with Generative Adversarial Networks

  title={Unfolding with Generative Adversarial Networks},
  author={Kaustuv Datta and Deepak Kar and D. Roy},
  journal={arXiv: Data Analysis, Statistics and Probability},
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 the data using a modified Generative Adversarial Network (MSGAN) is presented here. Applied to various distributions, it is demonstrated to perform at par with, or better than, currently used methods. 

Figures from this paper

How to GAN event subtraction

Generative adversarial networks are employed to produce new event samples with a phase space distribution corresponding to added or subtracted input samples and can be used to subtract background events or to include non-local collinear subtraction events at the level of unweighted 4-vector events.

Optimizing observables with machine learning for better unfolding

It is shown that using deep learning to define detector-level observables has the capability to improve the measurement when combined with standard unfolding methods.

LHC analysis-specific datasets with Generative Adversarial Networks

It is shown how substantial performance improvements and convergence speed-up can be obtained by including regression terms in the loss function of the generator, and an objective criterion is developed to assess the geenrator performance in a quantitative way.

The use of Generative Adversarial Networks to characterise new physics in multi-lepton final states at the LHC

Following the emergence of multi-lepton anomalies at the LHC, GANs are applied for the generation of di-leptons final states in association with b-quarks at theLHC with good agreement between the MC events and the WGAN-GP events.

Lund jet images from generative and cycle-consistent adversarial networks

A generative model to simulate radiation patterns within a jet using the Lund jet plane is introduced and a mapping can be created between different categories of jets, and this method is used to retroactively change simulation settings or the underlying process on an existing sample.

Hyperparameter Optimization of Generative Adversarial Network Models for High-Energy Physics Simulations

It is shown that given proper hyperparameter tuning, GANs that provide high-quality approximations of the desired quantities can be tuned and provided with guidelines for how to go about GAN architecture tuning using the analysis tools in Hyppo.

Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)

The new Feature-Augmented and Transformed GAN (FAT-GAN) is able to faithfully reproduce the distribution of final state electron momenta in inclusive electron scattering, without the need for input derived from domain-based theoretical assumptions.

GANplifying event samples

A critical question concerning generative networks applied to event generation in particle physics is if the generated events add statistical precision beyond the training sample. We show for a

Latent Space Refinement for Deep Generative Models

This work demonstrates how latent space refinement via iterated generative modeling can circumvent topological obstructions and improve precision and shows how this methodology also applies to cases were the target model is non-differentiable and has many internal latent dimensions which must be marginalized over before refinement.

DCTRGAN: improving the precision of generative models with reweighting

A post-hoc correction to deep generative models to further improve their fidelity, based on the Deep neural networks using the Classification for Tuning and Reweighting (DCTR) protocol is introduced.



Fast and Accurate Simulation of Particle Detectors Using Generative Adversarial Networks

It is shown that deep neural networks can achieve high fidelity in this task, while attaining a speed increase of several orders of magnitude with respect to traditional algorithms.

Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis

A simple architecture is proposed that learns to produce realistic radiation patterns from simulated high energy particle collisions and sheds light on limitations, and provides a novel empirical validation of image quality and validity of GAN-produced simulations of the natural world.

Accelerating Science with Generative Adversarial Networks: An Application to 3D Particle Showers in Multilayer Calorimeters.

A deep neural network-based generative model is introduced to enable high-fidelity, fast, electromagnetic calorimeter simulation and opens the door to a new era of fast simulation that could save significant computing time and disk space, while extending the reach of physics searches and precision measurements at the LHC and beyond.

Machine learning as an instrument for data unfolding

A method for correcting for detector smearing effects using machine learning techniques is presented and can use more than one reconstructed variable to infere the value of the unsmeared quantity on event by event basis.

GAN-D: Generative adversarial networks for image deconvolution

The loss function of the generator of GAN-D which combines mean square error of network output and ground-truth images to traditional adversarial loss of G AN is devised to produce more high-quality images than the original model structured with a single convolutional neural network.

Improved Techniques for Training GANs

This work focuses on two applications of GANs: semi-supervised learning, and the generation of images that humans find visually realistic, and presents ImageNet samples with unprecedented resolution and shows that the methods enable the model to learn recognizable features of ImageNet classes.

Controlling Physical Attributes in GAN-Accelerated Simulation of Electromagnetic Calorimeters

An auxiliary task is introduced to the training of a Generative Adversarial Network on particle showers in a multi-layer electromagnetic calorimeter, which allows the model to learn an attribute-aware conditioning mechanism.

CaloGAN: Simulating 3D High Energy Particle Showers in Multi-Layer Electromagnetic Calorimeters with Generative Adversarial Networks

CaloGAN, a new fast simulation technique based on generative adversarial networks (GANs) is introduced, which is applied to the modeling of electromagnetic showers in a longitudinally segmented calorimeter and achieves speedup factors comparable to or better than existing full simulation techniques.


Machine learning approach to inverse problem and unfolding procedure

A procedure for unfolding the true distribution from experimental data is presented. Machine learning method are applied for the identiflcation an apparatus function and solving inverse problem