Sebastiano Aiello, Arnauld Albert, Sergio Alves Garre, Zineb Aly, Fabrizio Ameli, Michel Andre, Giorgos Androulakis, Marco Anghinolfi, Mancia Anguita, Gisela Anton, Miquel Ardid, Julien Aublin, Christos Bagatelas, Giancarlo Barbarino, Bruny Baret, Suzan Basegmez du Pree, Meriem Bendahman, Edward Berbee, Vincent Bertin, Simone Biagi, Andrea Biagioni, Matthias Bissinger, Markus Boettcher, Jihad Boumaaza, Mohammed Bouta, Mieke Bouwhuis, Cristiano Bozza, Horea Branzas, Ronald Bruijn, Jürgen Brunner, Ernst-Jan Buis, Raffaele Buompane, Jose Busto, Barbara Caiffi, David Calvo, Antonio Capone, Víctor Carretero, Paolo Castaldi, Silvia Celli, Mohamed Chabab, Nhan Chau, Andrew Chen, Silvio Cherubini, Vitaliano Chiarella, Tommaso Chiarusi, Marco Circella, Rosanna Cocimano, Joao Coelho, Alexis Coleiro, Marta Colomer Molla, Rosa Coniglione, Paschal Coyle, Alexandre Creusot, Giacomo Cuttone, Antonio D'Onofrio, Richard Dallier, Maarten de Jong, Paul de Jong, Mauro De Palma, Gwenhaël de Wasseige, Els de Wolf, Irene Di Palma, Antonio Diaz, Dídac Diego-Tortosa, Carla Distefano, Alba Domi, Roberto Donà, Corinne Donzaud, Damien Dornic, Manuel Dörr, Doriane Drouhin, Thomas Eberl, Imad El Bojaddaini, Dominik Elsaesser, Alexander Enzenhöfer, Paolo Fermani, Giovanna Ferrara, Miroslav Filipovic, Francesco Filippini, Luigi Antonio Fusco, Omar Gabella, Tamas Gal, Alfonso Andres Garcia Soto, Fabio Garufi, Yoann Gatelet, Nicole Geißelbrecht, Lucio Gialanella, Emidio Giorgio, Sara Rebecca Gozzini, Rodrigo Gracia, Kay Graf, Dario Grasso, Giuseppe Grella, Carlo Guidi, Steffen Hallmann, Hassane Hamdaoui, Aart Heijboer, Amar Hekalo, Juan-Jose Hernandez-Rey, Jannik Hofestädt, Feifei Huang, Walid Idrissi Ibnsalih, Giulia Illuminati, Clancy James, Bouke Jisse Jung, Matthias Kadler, Piotr Kalaczyński, Oleg Kalekin, Uli Katz, Nafis Rezwan Khan Chowdhury, Giorgi Kistauri, Els Koffeman, Paul Kooijman, Antoine Kouchner, Michael Kreter, Vladimir Kulikovskiy, Robert Lahmann, Giuseppina Larosa, Remy Le Breton, Ornella Leonardi, Francesco Leone, Emanuele Leonora, Giuseppe Levi, Massimiliano Lincetto, Miles Lindsey Clark, Thomas Lipreau, Alessandro Lonardo, Fabio Longhitano, Daniel Lopez Coto, Lukas Maderer, Jerzy Mańczak, Karl Mannheim, Annarita Margiotta, Antonio Marinelli, Christos Markou, Lilian Martin, Juan Antonio Martínez-Mora, Agnese Martini, Fabio Marzaioli, Stefano Mastroianni, Safaa Mazzou, Karel Melis, Gennaro Miele, Pasquale Migliozzi, Emilio Migneco, Piotr Mijakowski, Luis Salvador Miranda Palacios, Carlos Maximiliano Mollo, Mauro Morganti, Michael Moser, Abdelilah Moussa, Rasa Muller, Mario Musumeci, Lodewijk Nauta, Sergio Navas, Carlo Alessandro Nicolau, Brían Ó Fearraigh, Mukharbek Organokov, Angelo Orlando, Gogita Papalashvili, Riccardo Papaleo, Cosimo Pastore, Alice Paun, Gabriela Emilia Pavalas, Carmelo Pellegrino, Mathieu Perrin-Terrin, Paolo Piattelli, Camiel Pieterse, Konstantinos Pikounis, Ofelia Pisanti, Chiara Poirè, Vlad Popa, Maarten Post, Thierry Pradier, Gerd Pühlhofer, Sara Pulvirenti, Omphile Rabyang, Fabrizio Raffaelli, Nunzio Randazzo, Antonio Rapicavoli, Soebur Razzaque, Diego Real, Stefan Reck, Giorgio Riccobene, Marc Richer, Stephane Rivoire, Alberto Rovelli, Francisco Salesa Greus, Dorothea Franziska Elisabeth Samtleben, Agustín Sánchez Losa, Matteo Sanguineti, Andrea Santangelo, Domenico Santonocito, Piera Sapienza, Jutta Schnabel, Jordan Seneca, Irene Sgura, Rezo Shanidze, Ankur Sharma, Francesco Simeone, Anna Sinopoulou, Bernardino Spisso, Maurizio Spurio, Dimitris Stavropoulos, Jos Steijger, Simona Maria Stellacci, Mauro Taiuti, Yahya Tayalati, Enrique Tenllado, Tarak Thakore, Steven Tingay, Ekaterini Tzamariudaki, Dimitrios Tzanetatos, Ad van den Berg, Frits van der Knaap, Daan van Eijk, Véronique Van Elewyck, Hans van Haren, Godefroy Vannoye, George Vasileiadis, Federico Versari, Salvatore Viola, Daniele Vivolo, Joern Wilms, Rafał Wojaczyński, Dmitry Zaborov, Sandra Zavatarelli, Angela Zegarelli, Daniele Zito, Juan-de-Dios Zornoza, Juan Zúñiga, Natalia Zywucka
The KM3NeT research infrastructure is currently under construction at two locations in the Mediterranean Sea. The KM3NeT/ORCA water-Cherenkov neutrino detector off the French coast will instrument several megatons of seawater with photosensors. Its main objective is the determination of the neutrino mass ordering. This work aims at demonstrating the general applicability of deep convolutional neural networks to neutrino telescopes, using simulated datasets for the KM3NeT/ORCA detector as an example. To this end, the networks are employed to achieve reconstruction and classification tasks that constitute an alternative to the analysis pipeline presented for KM3NeT/ORCA in the KM3NeT Letter of Intent. They are used to infer event reconstruction estimates for the energy, the direction, and the interaction point of incident neutrinos. The spatial distribution of Cherenkov light generated by charged particles induced in neutrino interactions is classified as shower- or track-like, and the main background processes associated with the detection of atmospheric neutrinos are recognized. Performance comparisons to machine-learning classification and maximum-likelihood reconstruction algorithms previously developed for KM3NeT/ORCA are provided. It is shown that this application of deep convolutional neural networks to simulated datasets for a large-volume neutrino telescope yields competitive reconstruction results and performance improvements with respect to classical approaches.
KM3NeT, a neutrino telescope currently under construction in the Mediterranean Sea, consists of a network of large-volume Cherenkov detectors. Its two different sites, ORCA and ARCA, are optimised for few GeV and TeV-PeV neutrino energies, respectively. This allows for studying a wide range of physics topics spanning from the determination of the neutrino mass hierarchy to the detection of neutrinos from astrophysical sources. Deep Learning techniques provide promising methods to analyse the signatures induced by charged particles traversing the detector. This document will cover a Deep Learni…
The KM3NeT collaboration is constructing two large neutrino detectors in the Mediterranean Sea: ARCA, located near Sicily and aiming at neutrino astronomy, and ORCA, located near Toulon and designed for neutrino oscillation studies. The two detectors, together, will have hundreds of Detection Units (DUs) with 18 Digital Optical Modules (DOMs) maintained vertical by buoyancy, forming a large 3D optical array for detecting the Cherenkov light produced after the neutrino interactions. To properly reconstruct the direction of the incoming neutrino, the position of the DOMs must be known precisely …
State of the art methods in astronomical image reconstruction rely on the resolution of a regularized or constrained optimization problem. Solving this problem can be computationally intensive and usually leads to a quadratic or at least superlinear complexity w.r.t. the number of pixels in the image. We investigate in this work the use of convolutional neural networks for image reconstruction in astronomy. With neural networks, the computationally intensive tasks is the training step, but the prediction step has a fixed complexity per pixel, i.e. a linear complexity. Numerical experiments sho…
The Cherenkov Telescope Array (CTA) will be the world's largest and most sensitive ground-based gamma-ray observatory in the energy range from a few tens of GeV to tens of TeV. The LST-1 prototype, currently in its commissioning phase, is the first of the four largest CTA telescopes, that will be built in the northern site of CTA in La Palma, Canary Islands, Spain. In this contribution, we present a full-image reconstruction method using a modified InceptionV3 deep convolutional neural network applied on non-parametrized shower images. We evaluate the performance of optimized networks on Monte…
Deep convolutional neural networks (DCNs) are a promising machine learning technique to reconstruct events recorded by imaging atmospheric Cherenkov telescopes (IACTs), but require optimization to reach full performance. One of the most pressing challenges is processing raw images captured by cameras made of hexagonal lattices of photo-multipliers, a common layout among IACT cameras which topologically differs from the square lattices conventionally expected, as their input data, by DCN models. Strategies directed to tackle this challenge range from the conversion of the hexagonal lattices ont…
Fourier-based wavefront sensors, such as the Pyramid Wavefront Sensor (PWFS), are the current preference for high contrast imaging due to their high sensitivity. However, these wavefront sensors have intrinsic nonlinearities that constrain the range where conventional linear reconstruction methods can be used to accurately estimate the incoming wavefront aberrations. We propose to use Convolutional Neural Networks (CNNs) for the nonlinear reconstruction of the wavefront sensor measurements. It is demonstrated that a CNN can be used to accurately reconstruct the nonlinearities in both simulatio…
We introduce a novel method for reconstructing the projected matter distributions of galaxy clusters with weak-lensing (WL) data based on convolutional neural network (CNN). Training datasets are generated with ray-tracing through cosmological simulations. We control the noise level of the galaxy shear catalog such that it mimics the typical properties of the existing ground-based WL observations of galaxy clusters. We find that the mass reconstruction by our multi-layered CNN with the architecture of alternating convolution and trans-convolution filters significantly outperforms the tradition…
A dramatic progress in the field of computer vision has been made in recent years by applying deep learning techniques. State-of-the-art performance in image recognition is thereby reached with Convolutional Neural Networks (CNNs). CNNs are a powerful class of artificial neural networks, characterized by requiring fewer connections and free parameters than traditional neural networks and exploiting spatial symmetries in the input data. Moreover, CNNs have the ability to automatically extract general characteristic features from data sets and create abstract data representations which can perfo…
The Cherenkov Telescope Array (CTA) is the future observatory for ground-based imaging atmospheric Cherenkov telescopes. Each telescope will provide a snapshot of gamma-ray induced particle showers by capturing the induced Cherenkov emission at ground level. The simulation of such events provides camera images that can be used as training data for convolutional neural networks (CNNs) to differentiate signals from background events and to determine the energy of the initial gamma-ray events. Pattern spectra are commonly used tools for image classification and provide the distributions of the si…
The Major Atmospheric Gamma Imaging Cherenkov (MAGIC) telescope system consists of two imaging atmospheric Cherenkov telescopes (IACTs) and is located on the Canary island of La Palma. IACTs are excellent tools to inspect the very-high-energy (few tens of GeV and above) gamma-ray sky by capturing images of the air showers, originated by the absorption of gamma rays and cosmic rays by the atmosphere, through the detection of Cherenkov photons emitted in the shower. One of the main factors determining the sensitivity of IACTs to gamma-ray sources, in general, is how well reconstructed the proper…
Signal-background separation and energy reconstruction of gamma rays using pattern spectra and convolutional neural networks for the Small-Sized Telescopes of the Cherenkov Telescope Array
The Major Atmospheric Gamma Imaging Cherenkov (MAGIC) telescope system is located on the Canary Island of La Palma and inspects the very high-energy (VHE, few tens of GeV and above) gamma-ray sky. MAGIC consists of two imaging atmospheric Cherenkov telescopes (IACTs), which capture images of the air showers originating from the absorption of gamma rays and cosmic rays by the atmosphere, through the detection of Cherenkov photons emitted in the shower. The sensitivity of IACTs to gamma-ray sources is mainly determined by the ability to reconstruct the properties (type, energy, and arrival direc…
