Jonas Glombitza

Application of Graph Networks to a wide-field Water-Cherenkov-based Gamma-Ray Observatory

Water-Cherenkov-based observatories form the high-duty-cycle and wide field of view backbone for observations of the gamma-ray sky at very high energies. In this work, we propose a deep learning application based on graph neural networks (GNNs) for background rejection and energy reconstruction and compare it to state-of-the-art approaches. We find that GNNs outperform hand-designed classification algorithms and observables in background rejection and find an improved energy resolution compared to template-based method.

Inference of the Mass Composition of Cosmic Rays with energies from 10^18.5 to 10^20 eV using the Pierre Auger Observatory and Deep Learning

We present measurements of the atmospheric depth of the shower maximum Xmax, inferred for the first time on an event-by-event level using the Surface Detector of the Pierre Auger Observatory. Using deep learning, we were able to extend measurements of the Xmax distributions up to energies of 100 EeV (10^20 eV), not yet revealed by current measurements, providing new insights into the mass composition of cosmic rays at extreme energies. Gaining a 10-fold increase in statistics compared to the Fluorescence Detector data, we find evidence that the rate of change of the average Xmax with the logarithm of energy features three breaks at 6.5±0.6 (stat)±1 (sys) EeV, 11±2 (stat)±1 (sys) EeV, and 31±5 (stat)±3 (sys) EeV, in the vicinity to the three prominent features (ankle, instep, suppression) of the cosmic-ray flux. The energy evolution of the mean and standard deviation of the measured Xmax distributions indicates that the mass composition becomes increasingly heavier and purer, thus being incompatible with a large fraction of light nuclei between 50 EeV and 100 EeV.

Measurement of the Depth of Maximum of Air-Shower Profiles with energies between 10^18.5 and 10^20 eV using the Surface Detector of the Pierre Auger Observatory and Deep Learning

We report an investigation of the mass composition of cosmic rays with energies from 3 to 100 EeV (1 EeV=1018 eV) using the distributions of the depth of shower maximum Xmax. The analysis relies on ∼50,000 events recorded by the Surface Detector of the Pierre Auger Observatory and a deep-learning-based reconstruction algorithm. Above energies of 5 EeV, the data set offers a 10-fold increase in statistics with respect to fluorescence measurements at the Observatory. After cross-calibration using the Fluorescence Detector, this enables the first measurement of the evolution of the mean and the standard deviation of the Xmax distributions up to 100 EeV. Our findings are threefold: (1.) The evolution of the mean logarithmic mass towards a heavier composition with increasing energy can be confirmed and is extended to 100 EeV. (2.) The evolution of the fluctuations of Xmax towards a heavier and purer composition with increasing energy can be confirmed with high statistics. We report a rather heavy composition and small fluctuations in Xmax at the highest energies. (3.) We find indications for a characteristic structure beyond a constant change in the mean logarithmic mass, featuring three breaks that are observed in proximity to the ankle, instep, and suppression features in the energy spectrum.

Ultra-Fast Generation of Air Shower Images for IACTs using Generative Adversarial Networks

For the analysis of data taken by Imaging Air Cherenkov Telescopes (IACTs), a large number of air shower simulations are needed to derive the instrument response. The simulations are very complex, involving computational and memory-intensive calculations, and are usually performed repeatedly for different observation intervals to take into account the varying optical sensitivity of the instrument. The use of generative models based on deep neural networks offers the prospect for memory-efficient storing of huge simulation libraries and cost-effective generation of a large number of simulations in an extremely short time. In this work, we use Wasserstein Generative Adversarial Networks to generate photon showers for an IACT equipped with the FlashCam design, which has more than 1,500 pixels. We demonstrate for the first time that the generated images have high fidelity with respect to low-level observables, the Hillas parameters, their physical properties, as well as their correlations. The found increase in generation speed in the order of 10^5 yields promising prospects for fast and memory-efficient simulations of air showers for IACTs.

Application of Graph Networks to background rejection in Imaging Air Cherenkov Telescopes

Imaging Air Cherenkov Telescopes (IACTs) are essential to ground-based observations of gamma rays in the GeV to TeV regime. One particular challenge of ground-based gamma-ray astronomy is an effective rejection of the hadronic background. We propose a new deep-learning-based algorithm for classifying images measured using single or multiple Imaging Air Cherenkov Telescopes by utilizing graph convolutional networks. For images cleaned of the light from the night sky, this allows for an efficient algorithm design that bypasses the challenge of sparse images in CNN-based approachs and find a promising performance with improvements to previous machine-learning and deep-learning-based methods.

Book: Deep Learning for Physics Research

A core principle of physics is knowledge gained from data. Thus, deep learning has instantly entered physics and may become a new paradigm in basic and applied research. This textbook addresses physics students and physicists who want to understand what deep learning actually means, and what is the potential for their own scientific projects. Being familiar with linear algebra and parameter optimization is sufficient to jump-start deep learning. Adopting a pragmatic approach, basic and advanced applications in physics research are described. Also offered are simple hands-on exercises for implementing deep networks for which python code and training data can be downloaded (see deeplearningphysics.org).

Ultra-High-Energy Cosmic Rays: The Intersection of the Cosmic and Energy Frontiers

The present white paper is submitted as part of the “Snowmass” process to help inform the long-term plans of the United States Department of Energy and the National Science Foundation for high-energy physics. It summarizes the science questions driving the Ultra-High-Energy Cosmic-Ray (UHECR) community and provides recommendations on the strategy to answer them in the next two decades.

Deep-learning based measurement of the mass composition of ultra-high energy cosmic rays using the surface detector of the Pierre Auger Observatory

Ph.D. thesis, RWTH Aachen University, 2021. December 2021

Shared Data and Algorithms for Deep Learning in Fundamental Physics

We introduce a collection of datasets from fundamental physics research ‐ including particle physics, astroparticle physics, and hadron- and nuclear physics ‐ for supervised machine learning studies. These datasets, containing hadronic top quarks, cosmic-ray induced air showers, phase transitions in hadronic matter, and generator-level histories, are made public to simplify future work on cross-disciplinary machine learning and transfer learning in fundamental physics. Based on these data, we present a simple yet flexible graph-based neural network architecture that can easily be applied to a wide range of supervised learning tasks in these domains. We show that our approach reaches performance close to state-of-the-art dedicated methods on all datasets. To simplify adaptation for various problems, we provide easy-to-follow instructions on how graph-based representations of data structures, relevant for fundamental physics, can be constructed and provide code implementations for several of them. Implementations are also provided for our proposed method and all reference algorithms.

Deep-Learning based Reconstruction of the Shower Maximum Xmax using the Water-Cherenkov Detectors of the Pierre Auger Observatory

The atmospheric depth of the air shower maximum Xmax is an observable commonly used for the determination of the nuclear mass composition of ultra-high energy cosmic rays. Direct measurements of Xmax are performed using observations of the longitudinal shower development with fluorescence telescopes. At the same time, several methods have been proposed for an indirect estimation of Xmax from the characteristics of the shower particles registered with surface detector arrays. In this paper, we present a deep neural network (DNN) for the estimation of Xmax. The reconstruction relies on the signals induced by shower particles in the ground based water-Cherenkov detectors of the Pierre Auger Observatory. The network architecture features recurrent long short-term memory layers to process the temporal structure of signals and hexagonal convolutions to exploit the symmetry of the surface detector array. We evaluate the performance of the network using air showers simulated with three different hadronic interaction models. Thereafter, we account for long-term detector effects and calibrate the reconstructed Xmax using fluorescence measurements. Finally, we show that the event-by-event resolution in the reconstruction of the shower maximum improves with increasing shower energy and reaches less than 25 g/cm² at energies above 2×10¹⁹ eV.

Adapting Simulations to Data at the Pierre Auger Observatory by refining Signal Traces using Cycle-Consistent Adversarial Networks

Identification of Patterns in Cosmic-Ray Arrival Directions using Dynamic Graph Convolutional Neural Networks

We present a new approach for the identification of ultra-high energy cosmic rays from sources using dynamic graph convolutional neural networks. These networks are designed to handle sparsely arranged objects and to exploit their short- and long-range correlations. Our method searches for patterns in the arrival directions of cosmic rays, which are expected to result from coherent deflections in cosmic magnetic fields. The network discriminates astrophysical scenarios with source signatures from those with only isotropically distributed cosmic rays and allows for the identification of cosmic rays that belong to a deflection pattern. We use simulated astrophysical scenarios where the source density is the only free parameter to show how density limits can be derived. We apply this method to a public data set from the AGASA Observatory.

Deep Neural Network for the Reconstruction of the Shower Maximum Xmax using the Water-Cherenkov Detectors of the Pierre Auger Observatory

Deep Learning based Algorithms in Astroparticle Physics

In recent years, great progress has been made in the fields of machine translation, image classification and speech recognition by using deep neural networks and associated techniques. Recently, the astroparticle physics community successfully adapted supervised learning algorithms for a wide range of tasks including background rejection, object reconstruction, track segmentation and the denoising of signals. Additionally, the first approaches towards fast simulations and simulation refinement indicate the huge potential of unsupervised learning for astroparticle physics. We summarize the latest results, discuss the algorithms and challenges and further illustrate the opportunities for the astrophysics community offered by deep learning based algorithms.

Proposal to investigate Composition Dependencies of Correlations between Ultra-High-Energy Cosmic Rays and SBG/AGN Source Candidates

Air-Shower Reconstruction at the Pierre Auger Observatory based on Deep Learning

The surface-detector array of the Pierre Auger Observatory measures the footprint of air showers induced by ultra-high energy cosmic rays. The reconstruction of event-by-event information sensitive to the cosmic-ray mass, is a challenging task and so far mainly based on fluorescence detector observations with their duty cycle of ≈ 15%. Recently, great progress has been made in multiple fields of machine learning using deep neural networks and associated techniques. Applying these new techniques to air-shower physics opens up possibilities for improved reconstruction, including an estimation of the cosmic-ray composition. In this contribution, we show that deep convolutional neural networks can be used for air-shower reconstruction, using surface-detector data. The focus of the machine learning algorithm is to reconstruct depths of shower maximum. In contrast to traditional reconstruction methods, the algorithm learns to extract the essential information from the signal and arrival-time distributions of the secondary particles. We present the neural-network architecture, describe the training, and assess the performance using simulated air showers.

Precise simulation of electromagnetic calorimeter showers using a Wasserstein Generative Adversarial Network

Simulations of particle showers in calorimeters are computationally time-consuming, as they have to reproduce both energy depositions and their considerable fluctuations. A new approach to ultra-fast simulations are generative models where all calorimeter energy depositions are generated simultaneously. We use GEANT4 simulations of an electron beam impinging on a multi-layer electromagnetic calorimeter for adversarial training of a generator network and a critic network guided by the Wasserstein distance. In most aspects, we observe that the generated calorimeter showers reach the level of showers as simulated with the GEANT4 program.

Generating and Refining Particle Detector Simulations Using the Wasserstein Distance in Adversarial Networks

We use adversarial frameworks together with the Wasserstein distance to generate or refine simulated detector data. The data reflect two-dimensional projections of spatially distributed signal patterns with a broad spectrum of applications. As an example, we use an observatory to detect cosmic ray-induced air showers with a ground-based array of particle detectors. First we investigate a method of generating detector patterns with variable signal strengths while constraining the primary particle energy. We then present a technique to refine simulated time traces of detectors to match corresponding data distributions. With this method we demonstrate that training a deep network with refined data-like signal traces leads to a more precise energy reconstruction of data events compared to training with the originally simulated traces.

A deep learning-based reconstruction of cosmic ray-induced air showers

We describe a method of reconstructing air showers induced by cosmic rays using deep learning techniques. We simulate an observatory consisting of ground-based particle detectors with fixed locations on a regular grid. The detector’s responses to traversing shower particles are signal amplitudes as a function of time, which provide information on transverse and longitudinal shower properties. In order to take advantage of convolutional network techniques specialized in local pattern recognition, we convert all information to the image-like grid of the detectors. In this way, multiple features, such as arrival times of the first particles and optimized characterizations of time traces, are processed by the network. The reconstruction quality of the cosmic ray arrival direction turns out to be competitive with an analytic reconstruction algorithm. The reconstructed shower direction, energy and shower depth show the expected improvement in resolution for higher cosmic ray energy.