-
Implicit Quantile Networks For Emulation in Jet Physics
Authors:
B. Kronheim,
A. Al Kadhim,
M. P. Kuchera,
H. B. Prosper,
R. Ramanujan
Abstract:
The ability to model and sample from conditional densities is important in many physics applications. Implicit quantile networks (IQN) have been successfully applied to this task in domains outside physics. In this work, we illustrate the potential of IQNs as components of emulators using the simulation of jets as an example. Specifically, we use an IQN to map jets described by their 4-momenta at…
▽ More
The ability to model and sample from conditional densities is important in many physics applications. Implicit quantile networks (IQN) have been successfully applied to this task in domains outside physics. In this work, we illustrate the potential of IQNs as components of emulators using the simulation of jets as an example. Specifically, we use an IQN to map jets described by their 4-momenta at the generation level to jets at the event reconstruction level. The conditional densities emulated by our model closely match those generated by $\texttt{Delphes}$, while also enabling faster jet simulation.
△ Less
Submitted 26 June, 2023;
originally announced June 2023.
-
Two-dimensional total absorption spectroscopy with conditional generative adversarial networks
Authors:
Cade Dembski,
Michelle P. Kuchera,
Sean Liddick,
Raghu Ramanujan,
Artemis Spyrou
Abstract:
We explore the use of machine learning techniques to remove the response of large volume $γ$-ray detectors from experimental spectra. Segmented $γ$-ray total absorption spectrometers (TAS) allow for the simultaneous measurement of individual $γ$-ray energy (E$_γ$) and total excitation energy (E$_x$). Analysis of TAS detector data is complicated by the fact that the E$_x$ and E$_γ$ quantities are c…
▽ More
We explore the use of machine learning techniques to remove the response of large volume $γ$-ray detectors from experimental spectra. Segmented $γ$-ray total absorption spectrometers (TAS) allow for the simultaneous measurement of individual $γ$-ray energy (E$_γ$) and total excitation energy (E$_x$). Analysis of TAS detector data is complicated by the fact that the E$_x$ and E$_γ$ quantities are correlated, and therefore, techniques that simply unfold using E$_x$ and E$_γ$ response functions independently are not as accurate. In this work, we investigate the use of conditional generative adversarial networks (cGANs) to simultaneously unfold $E_{x}$ and $E_γ$ data in TAS detectors. Specifically, we employ a \texttt{Pix2Pix} cGAN, a generative modeling technique based on recent advances in deep learning, to treat \rawmatrix~ matrix unfolding as an image-to-image translation problem. We present results for simulated and experimental matrices of single-$γ$ and double-$γ$ decay cascades. Our model demonstrates characterization capabilities within detector resolution limits for upwards of 93% of simulated test cases.
△ Less
Submitted 14 December, 2023; v1 submitted 23 June, 2022;
originally announced June 2022.
-
Machine Learning in Nuclear Physics
Authors:
Amber Boehnlein,
Markus Diefenthaler,
Cristiano Fanelli,
Morten Hjorth-Jensen,
Tanja Horn,
Michelle P. Kuchera,
Dean Lee,
Witold Nazarewicz,
Kostas Orginos,
Peter Ostroumov,
Long-Gang Pang,
Alan Poon,
Nobuo Sato,
Malachi Schram,
Alexander Scheinker,
Michael S. Smith,
Xin-Nian Wang,
Veronique Ziegler
Abstract:
Advances in machine learning methods provide tools that have broad applicability in scientific research. These techniques are being applied across the diversity of nuclear physics research topics, leading to advances that will facilitate scientific discoveries and societal applications.
This Review gives a snapshot of nuclear physics research which has been transformed by machine learning techni…
▽ More
Advances in machine learning methods provide tools that have broad applicability in scientific research. These techniques are being applied across the diversity of nuclear physics research topics, leading to advances that will facilitate scientific discoveries and societal applications.
This Review gives a snapshot of nuclear physics research which has been transformed by machine learning techniques.
△ Less
Submitted 2 May, 2022; v1 submitted 4 December, 2021;
originally announced December 2021.
-
Implicit Quantile Neural Networks for Jet Simulation and Correction
Authors:
Braden Kronheim,
Michelle P. Kuchera,
Harrison B. Prosper,
Raghuram Ramanujan
Abstract:
Reliable modeling of conditional densities is important for quantitative scientific fields such as particle physics. In domains outside physics, implicit quantile neural networks (IQN) have been shown to provide accurate models of conditional densities. We present a successful application of IQNs to jet simulation and correction using the tools and simulated data from the Compact Muon Solenoid (CM…
▽ More
Reliable modeling of conditional densities is important for quantitative scientific fields such as particle physics. In domains outside physics, implicit quantile neural networks (IQN) have been shown to provide accurate models of conditional densities. We present a successful application of IQNs to jet simulation and correction using the tools and simulated data from the Compact Muon Solenoid (CMS) Open Data portal.
△ Less
Submitted 22 November, 2021;
originally announced November 2021.
-
Machine learning-based event generator for electron-proton scattering
Authors:
Y. Alanazi,
P. Ambrozewicz,
M. Battaglieri,
A. N. Hiller Blin,
M. P. Kuchera,
Y. Li,
T. Liu,
R. E. McClellan,
W. Melnitchouk,
E. Pritchard,
M. Robertson,
N. Sato,
R. Strauss,
L. Velasco
Abstract:
We present a new machine learning-based Monte Carlo event generator using generative adversarial networks (GANs) that can be trained with calibrated detector simulations to construct a vertex-level event generator free of theoretical assumptions about femtometer scale physics. Our framework includes a GAN-based detector folding as a fast-surrogate model that mimics detector simulators. The framewo…
▽ More
We present a new machine learning-based Monte Carlo event generator using generative adversarial networks (GANs) that can be trained with calibrated detector simulations to construct a vertex-level event generator free of theoretical assumptions about femtometer scale physics. Our framework includes a GAN-based detector folding as a fast-surrogate model that mimics detector simulators. The framework is tested and validated on simulated inclusive deep-inelastic scattering data along with existing parametrizations for detector simulation, with uncertainty quantification based on a statistical bootstrap** technique. Our results provide for the first time a realistic proof-of-concept to mitigate theory bias in inferring vertex-level event distributions needed to reconstruct physical observables.
△ Less
Submitted 19 April, 2022; v1 submitted 6 August, 2020;
originally announced August 2020.
-
Unsupervised Learning for Identifying Events in Active Target Experiments
Authors:
Robert Solli,
Daniel Bazin,
Michelle P. Kuchera,
Ryan R. Strauss,
Morten Hjorth-Jensen
Abstract:
This article presents novel applications of unsupervised machine learning methods to the problem of event separation in an active target detector, the Active-Target Time Projection Chamber (AT-TPC). The overarching goal is to group similar events in the early stages of the data analysis, thereby improving efficiency by limiting the computationally expensive processing of unnecessary events. The ap…
▽ More
This article presents novel applications of unsupervised machine learning methods to the problem of event separation in an active target detector, the Active-Target Time Projection Chamber (AT-TPC). The overarching goal is to group similar events in the early stages of the data analysis, thereby improving efficiency by limiting the computationally expensive processing of unnecessary events. The application of unsupervised clustering algorithms to the analysis of two-dimensional projections of particle tracks from a resonant proton scattering experiment on $^{46}$Ar is introduced. We explore the performance of autoencoder neural networks and a pre-trained VGG16 convolutional neural network. We study clustering performance on both data from a simulated $^{46}$Ar experiment, and real events from the AT-TPC detector. We find that a $k$-means algorithm applied to simulated data in the VGG16 latent space forms almost perfect clusters. Additionally, the VGG16+$k$-means approach finds high purity clusters of proton events for real experimental data. We also explore the application of clustering the latent space of autoencoder neural networks for event separation. While these networks show strong performance, they suffer from high variability in their results.
△ Less
Submitted 13 March, 2021; v1 submitted 6 August, 2020;
originally announced August 2020.
-
Modeling Magnetohydrodynamic Equilibrium in Magnetars with Applications to Continuous Gravitational Wave Production
Authors:
Samuel G. Frederick,
Kristen L. Thompson,
Michelle P. Kuchera
Abstract:
Possessing the strongest magnetic fields in the Universe, magnetars mark an extremum of physical phenomena. The strength of their magnetic fields is sufficient to deform the shape of the stellar body, and when the rotational and magnetic axes are not aligned, these deformations lead to the production of gravitational waves (GWs) via a time-varying quadrupole moment. Such gravitational radiation di…
▽ More
Possessing the strongest magnetic fields in the Universe, magnetars mark an extremum of physical phenomena. The strength of their magnetic fields is sufficient to deform the shape of the stellar body, and when the rotational and magnetic axes are not aligned, these deformations lead to the production of gravitational waves (GWs) via a time-varying quadrupole moment. Such gravitational radiation differs from signals presently detectable by the Laser Interferometer Gravitational-Wave Observatory. These signals are continuous rather than the momentary 'chirp' waveforms produced by binary systems during the phases of inspiral, merger, and ringdown. We evaluate the sensitivity requirements of future iterations of GW detectors to continuous GW signals resulting from magnetars. Here, we construct a computational model for magnetar stellar structure with strong internal magnetic fields. We implement an n = 1 polytropic equation of state (EOS) and adopt a mixed poloidal and toroidal magnetic field model constrained by the choice of EOS. We utilize fiducial values for magnetar magnetic field strength and various stellar physical attributes. Via computational simulation, we measure the deformation of magnetar stellar structure to determine upper bounds on the strength of continuous GWs formed as a result of these deformations inducing non-axisymmetric rotation. We compute predictions of upper limit GW strain values for sources in the McGill Magnetar Catalog, an index of all detected magnetars.
△ Less
Submitted 14 January, 2021; v1 submitted 6 February, 2020;
originally announced February 2020.
-
Simulation of electron-proton scattering events by a Feature-Augmented and Transformed Generative Adversarial Network (FAT-GAN)
Authors:
Yasir Alanazi,
N. Sato,
Tianbo Liu,
W. Melnitchouk,
Pawel Ambrozewicz,
Florian Hauenstein,
Michelle P. Kuchera,
Evan Pritchard,
Michael Robertson,
Ryan Strauss,
Luisa Velasco,
Yaohang Li
Abstract:
We apply generative adversarial network (GAN) technology to build an event generator that simulates particle production in electron-proton scattering that is free of theoretical assumptions about underlying particle dynamics. The difficulty of efficiently training a GAN event simulator lies in learning the complicated patterns of the distributions of the particles physical properties. We develop a…
▽ More
We apply generative adversarial network (GAN) technology to build an event generator that simulates particle production in electron-proton scattering that is free of theoretical assumptions about underlying particle dynamics. The difficulty of efficiently training a GAN event simulator lies in learning the complicated patterns of the distributions of the particles physical properties. We develop a GAN that selects a set of transformed features from particle momenta that can be generated easily by the generator, and uses these to produce a set of augmented features that improve the sensitivity of the discriminator. 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. The developed technology can play a significant role in boosting the science of existing and future accelerator facilities, such as the Electron-Ion Collider.
△ Less
Submitted 27 May, 2021; v1 submitted 29 January, 2020;
originally announced January 2020.
-
Machine Learning Methods for Track Classification in the AT-TPC
Authors:
Michelle P. Kuchera,
Raghuram Ramanujan,
Jack Z. Taylor,
Ryan R. Strauss,
Daniel Bazin,
Joshua Bradt,
Ruiming Chen
Abstract:
We evaluate machine learning methods for event classification in the Active-Target Time Projection Chamber detector at the National Superconducting Cyclotron Laboratory (NSCL) at Michigan State University. An automated method to single out the desired reaction product would result in more accurate physics results as well as a faster analysis process. Binary and multi-class classification methods w…
▽ More
We evaluate machine learning methods for event classification in the Active-Target Time Projection Chamber detector at the National Superconducting Cyclotron Laboratory (NSCL) at Michigan State University. An automated method to single out the desired reaction product would result in more accurate physics results as well as a faster analysis process. Binary and multi-class classification methods were tested on data produced by the $^{46}$Ar(p,p) experiment run at the NSCL in September 2015. We found a Convolutional Neural Network to be the most successful classifier of proton scattering events for transfer learning. Results from this investigation and recommendations for event classification in future experiments are presented.
△ Less
Submitted 29 April, 2019; v1 submitted 21 October, 2018;
originally announced October 2018.
