Valentina La Parola
Research Subjects
Research Interests
Valentina La Parola has a total of 1000 co-authors. Here are the top co-authors by number of shared publications:
- Giancarlo Cusumano
INAF -Istituto di Astrofisica Spaziale e Fisica Cosmica di Palermo Via U. La Malfa 153 90146 Palermo Italy
104 shared publications - Patrizia Romano
INAF – Osservatorio Astronomico di Brera, via Brera 28, 20121 Milano, Italy
60 shared publications - Gianpiero Tagliaferri
INAF -Osservatorio Astronomico di Brera via E. Bianchi 46 I-23807 Merate Italy
54 shared publications - Sergio Campana
INAF – Osservatorio Astronomico di Brera, Merate, Italy
51 shared publications - Vanessa Mangano
Department of Astronomy and Astrophysics, Pennsylvania State University, University Park, PA 16802, USA
49 shared publications G. Chincarini
Università degli studi di Milano-Bicocca, Dipartimento di Fisica, Piazza delle Scienze 3, 20126 Milan, Italy
47 shared publications
Latest Publications
- Open Access
- 23 September 2025
- Published by Sissa Medialab Srl
- Vol. 501, 717
New time-based parameters to enhance gamma-ray and hadron discrimination for the ASTRI Mini-Array
The ASTRI Mini-Array is an international project to build and operate an array of nine 4-m diameter Imaging Atmospheric Cherenkov Telescopes (IACT) at the Observatorio del Teide (Tenerife, Spain). The array has been designed to perform deep galactic and extragalactic gamma-ray sky observations in the 1-200 TeV energy range. As of today, the first telescope ASTRI-1 is fully operative, and in its commissioning phase. In the effort of improving the sensitivity of the array, particularly in discriminating between gamma-ray- and hadron-induced showers, we are investigating whether the temporal evolution of the shower images, as recorded by the pixel time tag, can provide additional discriminatory power beyond the traditional morphological parameters. We have developed and tested a set of parameters and through extensive testing, we have identified a subset of them that exhibit good discriminatory efficacy. In our preliminary analysis, combining the time parameters with the standard morphological ones has demonstrated a significant improvement in hadron rejection, especially at the lower end of the energy detection range (below 1 TeV), and a definite boost in the single telescope performance. Here we show also the distributions of this new set of parameters in the ASTRI-1 data taken in the November 2024 – February 2025 campaign and how they compare with the simulations.
- 23 September 2025
- Published by Sissa Medialab Srl
- Vol. 501, 717
Open AccessNew time-based parameters to enhance gamma-ray and hadron discrimination for the ASTRI Mini-Array
The ASTRI Mini-Array is an international project to build and operate an array of nine 4-m diameter Imaging Atmospheric Cherenkov Telescopes (IACT) at the Observatorio del Teide (Tenerife, Spain). The array has been designed to perform deep galactic and extragalactic gamma-ray sky observations in the 1-200 TeV energy range. As of today, the first telescope ASTRI-1 is fully operative, and in its commissioning phase. In the effort of improving the sensitivity of the array, particularly in discriminating between gamma-ray- and hadron-induced showers, we are investigating whether the temporal evolution of the shower images, as recorded by the pixel time tag, can provide additional discriminatory power beyond the traditional morphological parameters. We have developed and tested a set of parameters and through extensive testing, we have identified a subset of them that exhibit good discriminatory efficacy. In our preliminary analysis, combining the time parameters with the standard morphological ones has demonstrated a significant improvement in hadron rejection, especially at the lower end of the energy detection range (below 1 TeV), and a definite boost in the single telescope performance. Here we show also the distributions of this new set of parameters in the ASTRI-1 data taken in the November 2024 – February 2025 campaign and how they compare with the simulations.
- Open Access
- 5 August 2025
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 700, A70
How the spin-phase variability of cyclotron lines shapes the pulsed fraction spectra: Insights from 4U 1538–52
Aims. We aim to study the energy-dependent pulse profile of the X-ray accreting pulsar 4U 1538–52 and its phase-dependent spectral variability, with a particular emphasis on the behavior around the cyclotron resonant scattering feature at Ecyc ∼ 21 keV. Methods. We analyzed all available NuSTAR observations of 4U 1538–52. We decomposed the energy-resolved pulse profiles into Fourier harmonics to study their energy dependence. Specifically, we computed the pulsed fraction spectra, cross-correlation, and lag spectra, identifying discontinuities and linking them to features in the phase-averaged spectra. We performed both phase-averaged and phase-resolved spectral analyses to probe spectral variability and its relation to pulse profile changes. Finally, we interpreted our findings based on a physical modeling of the energy- and angle-dependent pulse profile emission, performing radiative transfer in a homogeneous slab-like atmosphere under conditions relevant to 4U 1538–52. The emission is projected onto the observer’s sky plane to derive the expected observables. Results. In contrast to the dips in pulsed fraction spectra observed in other sources (e.g., Her X-1), we find a broad bump near the cyclotron resonance energy in 4U 1538–52. This increase is driven primarily by phase-dependent spectral variability, especially by strong variations in cyclotron line depth across different phase intervals. We interpreted the observed contrast between dips and bumps in various sources as arising from phase-dependent variations of cyclotron line depth relative to the phase-modulated flux. We modeled the X-ray emission from an accreting neutron star and found that our simulations indicate high values of both the observer’s inclination and the magnetic obliquity, along with a ∼10 − 15° asymmetry between the locations of the magnetic poles. Assuming this geometry, we were able to adequately reproduce the observed pulse profiles and introduce general trends in the observables resulting from the system’s geometry.
- 5 August 2025
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 700, A70
Open AccessHow the spin-phase variability of cyclotron lines shapes the pulsed fraction spectra: Insights from 4U 1538–52
Aims. We aim to study the energy-dependent pulse profile of the X-ray accreting pulsar 4U 1538–52 and its phase-dependent spectral variability, with a particular emphasis on the behavior around the cyclotron resonant scattering feature at Ecyc ∼ 21 keV. Methods. We analyzed all available NuSTAR observations of 4U 1538–52. We decomposed the energy-resolved pulse profiles into Fourier harmonics to study their energy dependence. Specifically, we computed the pulsed fraction spectra, cross-correlation, and lag spectra, identifying discontinuities and linking them to features in the phase-averaged spectra. We performed both phase-averaged and phase-resolved spectral analyses to probe spectral variability and its relation to pulse profile changes. Finally, we interpreted our findings based on a physical modeling of the energy- and angle-dependent pulse profile emission, performing radiative transfer in a homogeneous slab-like atmosphere under conditions relevant to 4U 1538–52. The emission is projected onto the observer’s sky plane to derive the expected observables. Results. In contrast to the dips in pulsed fraction spectra observed in other sources (e.g., Her X-1), we find a broad bump near the cyclotron resonance energy in 4U 1538–52. This increase is driven primarily by phase-dependent spectral variability, especially by strong variations in cyclotron line depth across different phase intervals. We interpreted the observed contrast between dips and bumps in various sources as arising from phase-dependent variations of cyclotron line depth relative to the phase-modulated flux. We modeled the X-ray emission from an accreting neutron star and found that our simulations indicate high values of both the observer’s inclination and the magnetic obliquity, along with a ∼10 − 15° asymmetry between the locations of the magnetic poles. Assuming this geometry, we were able to adequately reproduce the observed pulse profiles and introduce general trends in the observables resulting from the system’s geometry.
- Open Access
How the spin-phase variability of cyclotron lines shapes the pulsed fraction spectra: insights from 4U 1538-52
We study the energy-dependent pulse profile of 4U 1538-52 and its phase-dependent spectral variability, with emphasis on the behavior around the cyclotron resonant scattering feature at around 21 keV. We analyze all available NuSTAR observations of 4U 1538-52. We decompose energy-resolved pulse profiles into Fourier harmonics to study their energy dependence. Specifically, we compute pulsed fraction spectra, cross-correlation and lag spectra, identifying discontinuities and linking them to features in the phase-averaged spectra. We perform phase-averaged and phase-resolved spectral analyses to probe spectral variability and its relation to pulse profile changes. Finally, we interpret our findings via physical modeling of energy- and angle-dependent pulse profile emission, performing radiative transfer in a homogeneous slab-like atmosphere under conditions relevant to 4U 1538-52. The emission is projected onto the observer's sky plane to derive expected observables. In contrast to the dips in pulsed fraction spectra observed in other sources (e.g., Her X-1), we find a broad bump near the cyclotron resonance energy in 4U 1538-52. This increase is driven primarily by phase-dependent spectral variability, especially by strong variations in cyclotron line depth across different phase intervals. We interpret the observed contrast between dips and bumps in various sources as arising from phase-dependent variations of cyclotron line depth relative to the phase-modulated flux. We model the X-ray emission from an accreting neutron star and find that our simulations indicate high values of both the observer's inclination and the magnetic obliquity, along with a 10-15 degrees asymmetry between the locations of the magnetic poles. Assuming this geometry, we satisfactorily reproduce the observed pulse profiles and introduce general trends in the observables resulting from the system's geometry.
Open AccessHow the spin-phase variability of cyclotron lines shapes the pulsed fraction spectra: insights from 4U 1538-52
We study the energy-dependent pulse profile of 4U 1538-52 and its phase-dependent spectral variability, with emphasis on the behavior around the cyclotron resonant scattering feature at around 21 keV. We analyze all available NuSTAR observations of 4U 1538-52. We decompose energy-resolved pulse profiles into Fourier harmonics to study their energy dependence. Specifically, we compute pulsed fraction spectra, cross-correlation and lag spectra, identifying discontinuities and linking them to features in the phase-averaged spectra. We perform phase-averaged and phase-resolved spectral analyses to probe spectral variability and its relation to pulse profile changes. Finally, we interpret our findings via physical modeling of energy- and angle-dependent pulse profile emission, performing radiative transfer in a homogeneous slab-like atmosphere under conditions relevant to 4U 1538-52. The emission is projected onto the observer's sky plane to derive expected observables. In contrast to the dips in pulsed fraction spectra observed in other sources (e.g., Her X-1), we find a broad bump near the cyclotron resonance energy in 4U 1538-52. This increase is driven primarily by phase-dependent spectral variability, especially by strong variations in cyclotron line depth across different phase intervals. We interpret the observed contrast between dips and bumps in various sources as arising from phase-dependent variations of cyclotron line depth relative to the phase-modulated flux. We model the X-ray emission from an accreting neutron star and find that our simulations indicate high values of both the observer's inclination and the magnetic obliquity, along with a 10-15 degrees asymmetry between the locations of the magnetic poles. Assuming this geometry, we satisfactorily reproduce the observed pulse profiles and introduce general trends in the observables resulting from the system's geometry.
- Open Access
- 1 April 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (7) , 3879
Machine Learning-Enhanced Discrimination of Gamma-Ray and Hadron Events Using Temporal Features: An ASTRI Mini-Array Analysis
Imaging Atmospheric Cherenkov Telescopes (IACTs) have revolutionized our understanding of the universe at very high energies (VHEs), enabling groundbreaking discoveries of extreme astrophysical phenomena. These instruments capture the brief flashes of Cherenkov light produced when VHE particles interact with Earth’s atmosphere, providing unique insights into cosmic accelerators and high-energy radiation sources. A fundamental challenge in IACT observations lies in distinguishing the rare gamma-ray signals from an overwhelming background of cosmic-ray events. For every gamma-ray photon detected from even the brightest sources, thousands of cosmic-ray-induced atmospheric showers trigger the telescopes. This profound signal-to-background imbalance necessitates sophisticated discrimination techniques that can effectively isolate genuine gamma-ray events while maintaining high rejection efficiency for cosmic-ray backgrounds. The most common method involves the parametrization of the morphological feature of the shower images. However, we know that gamma-ray and hadron showers also differ in their time evolution. Here, we describe how the pixel time tags (i.e., the record of when each camera pixel is lit up by the incoming shower) can help in the discrimination between photonic and hadronic showers, with a focus on the ASTRI Mini-Array Cherenkov Event Reconstruction. Our methodology employs a Random Forest classifier with optimized hyperparameters, trained on a balanced dataset of gamma and hadron events. The model incorporates feature importance analysis to select the most discriminating temporal parameters from a comprehensive set of time-based features. This machine learning approach enables effective integration of both morphological and temporal information, resulting in improved classification performance, especially at lower energies.
- 1 April 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (7) , 3879
Open AccessMachine Learning-Enhanced Discrimination of Gamma-Ray and Hadron Events Using Temporal Features: An ASTRI Mini-Array Analysis
Imaging Atmospheric Cherenkov Telescopes (IACTs) have revolutionized our understanding of the universe at very high energies (VHEs), enabling groundbreaking discoveries of extreme astrophysical phenomena. These instruments capture the brief flashes of Cherenkov light produced when VHE particles interact with Earth’s atmosphere, providing unique insights into cosmic accelerators and high-energy radiation sources. A fundamental challenge in IACT observations lies in distinguishing the rare gamma-ray signals from an overwhelming background of cosmic-ray events. For every gamma-ray photon detected from even the brightest sources, thousands of cosmic-ray-induced atmospheric showers trigger the telescopes. This profound signal-to-background imbalance necessitates sophisticated discrimination techniques that can effectively isolate genuine gamma-ray events while maintaining high rejection efficiency for cosmic-ray backgrounds. The most common method involves the parametrization of the morphological feature of the shower images. However, we know that gamma-ray and hadron showers also differ in their time evolution. Here, we describe how the pixel time tags (i.e., the record of when each camera pixel is lit up by the incoming shower) can help in the discrimination between photonic and hadronic showers, with a focus on the ASTRI Mini-Array Cherenkov Event Reconstruction. Our methodology employs a Random Forest classifier with optimized hyperparameters, trained on a balanced dataset of gamma and hadron events. The model incorporates feature importance analysis to select the most discriminating temporal parameters from a comprehensive set of time-based features. This machine learning approach enables effective integration of both morphological and temporal information, resulting in improved classification performance, especially at lower energies.
- Open Access
- 24 February 2025
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 694, A316
Energy-resolved pulse profile changes in V 0332+53: Indications of wings in the cyclotron absorption line profile
Aims. We aim to investigate the energy-resolved pulse profile changes of the accreting X-ray pulsar V 0332+53 focusing in the cyclotron line energy range, using the full set of available NuSTAR observations. Methods. We applied a tailored pipeline to study the energy dependence of the pulse profiles and to build the pulsed fraction spectra (PFS) for the different observations. We also studied the profile changes using cross-correlation and lag spectra. We re-analysed the energy spectra to search for links between the local features observed in the PFS and spectral emission components associated with the shape of the fundamental cyclotron line. Results. In the PFS data, with sufficiently high statistics, we observe a consistent behaviour around the cyclotron line energy. Specifically, two Gaussian-shaped features appear symmetrically on either side of the putative cyclotron line. These features exhibit minimal variation with source luminosity, and their peak positions consistently remain on the left and right of the cyclotron line energy. Associated with the cyclotron line-forming region, we interpret them as evidence for the resonant cyclotron absorption line wings, as predicted by theoretical models of how the cyclotron line profile should appear along the observer’s line of sight. A phase-resolved analysis of the pulse in the energy bands surrounding these features enables us to determine both the spectral shape and the intensity of the photons responsible for these peaks in the PFS. Assuming these features correspond to a spectral component, we used their shapes as priors for the corresponding emission components, finding a statistically satisfactory description of the spectra. To explain these results, we propose that our line of sight is close to the direction of the spin axis, while the magnetic axis is likely orthogonal to it.
- 24 February 2025
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 694, A316
Open AccessEnergy-resolved pulse profile changes in V 0332+53: Indications of wings in the cyclotron absorption line profile
Aims. We aim to investigate the energy-resolved pulse profile changes of the accreting X-ray pulsar V 0332+53 focusing in the cyclotron line energy range, using the full set of available NuSTAR observations. Methods. We applied a tailored pipeline to study the energy dependence of the pulse profiles and to build the pulsed fraction spectra (PFS) for the different observations. We also studied the profile changes using cross-correlation and lag spectra. We re-analysed the energy spectra to search for links between the local features observed in the PFS and spectral emission components associated with the shape of the fundamental cyclotron line. Results. In the PFS data, with sufficiently high statistics, we observe a consistent behaviour around the cyclotron line energy. Specifically, two Gaussian-shaped features appear symmetrically on either side of the putative cyclotron line. These features exhibit minimal variation with source luminosity, and their peak positions consistently remain on the left and right of the cyclotron line energy. Associated with the cyclotron line-forming region, we interpret them as evidence for the resonant cyclotron absorption line wings, as predicted by theoretical models of how the cyclotron line profile should appear along the observer’s line of sight. A phase-resolved analysis of the pulse in the energy bands surrounding these features enables us to determine both the spectral shape and the intensity of the photons responsible for these peaks in the PFS. Assuming these features correspond to a spectral component, we used their shapes as priors for the corresponding emission components, finding a statistically satisfactory description of the spectra. To explain these results, we propose that our line of sight is close to the direction of the spin axis, while the magnetic axis is likely orthogonal to it.
- Open Access
- 16 December 2024
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 692, A234
Tracking the long-term timing accuracy of the X-Ray Telescope on board the Neil Gehrels Swift Observatory
Context. The Neil Gehrels Swift Observatory has been operational since November 2004. Its X-ray Telescope (XRT), operating in the 0.3–10.0 keV range, is designed to provide detailed position, timing, and spectroscopic information. Aims. The calibration procedure for assessing the absolute timing accuracy of XRT was described in a previous paper. Here we update the past analysis using the complete data set of the Crab pulsar observations up to October 2022 and using a new version of the data-processing software package that includes corrections to several issues that could have affected the previous results. Methods. We evaluate the accuracy of the Crab pulse period determination using the folding technique and the pulse-phase analysis and compare our results with the values derived from radio observations. We also check the absolute time reconstruction, measuring the phase position of the main peak in the Crab profile and comparing it with the value reported in the literature, which is based on Rossi X-Ray Timing Explorer (RXTE) observations. Results. We find that the accuracy in period determination for the Crab pulsar is of the order of a few picoseconds for the observations with the largest data time span. The absolute time reconstruction, measured using the position of the main pulse peak, shows that the main peak precedes the phase of the position reported in the literature for RXTE by ~263 µs on average. This corresponds to 0.982 in phase, with an observed dispersion of ±0.02 in phase values. We also find that observations very close in time (down to ~1 day separation) show a significant variation in absolute phase.
- 16 December 2024
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 692, A234
Open AccessTracking the long-term timing accuracy of the X-Ray Telescope on board the Neil Gehrels Swift Observatory
Context. The Neil Gehrels Swift Observatory has been operational since November 2004. Its X-ray Telescope (XRT), operating in the 0.3–10.0 keV range, is designed to provide detailed position, timing, and spectroscopic information. Aims. The calibration procedure for assessing the absolute timing accuracy of XRT was described in a previous paper. Here we update the past analysis using the complete data set of the Crab pulsar observations up to October 2022 and using a new version of the data-processing software package that includes corrections to several issues that could have affected the previous results. Methods. We evaluate the accuracy of the Crab pulse period determination using the folding technique and the pulse-phase analysis and compare our results with the values derived from radio observations. We also check the absolute time reconstruction, measuring the phase position of the main peak in the Crab profile and comparing it with the value reported in the literature, which is based on Rossi X-Ray Timing Explorer (RXTE) observations. Results. We find that the accuracy in period determination for the Crab pulsar is of the order of a few picoseconds for the observations with the largest data time span. The absolute time reconstruction, measured using the position of the main pulse peak, shows that the main peak precedes the phase of the position reported in the literature for RXTE by ~263 µs on average. This corresponds to 0.982 in phase, with an observed dispersion of ±0.02 in phase values. We also find that observations very close in time (down to ~1 day separation) show a significant variation in absolute phase.
- Open Access
- 25 July 2023
- Published by Sissa Medialab Srl
- Vol. 444, 597
The Stereo Event Builder software system of the ASTRI Mini-Array
The Stereo Event Builder (SEB) software system of the ASTRI Mini-Array is the part of the off-line data processing chain that is responsible for identifying single and stereo Cherenkov events. The ASTRI Mini-Array is an international project led by the Italian National Institute for Astrophysics (INAF) which is in the process of deploying nine Imaging Atmospheric Cherenkov Telescopes (IACTs) of the 4-m class at the Observatorio del Teide in Tenerife (Spain). The project is designed to detect very high-energy gamma rays up to the multi-TeV scale. In the ASTRI Mini-Array operation concept, the expected performance is based on the stereoscopic technique, i.e. the detection of the same atmospheric shower event with two or more telescopes. However all single-telescope events are acquired independently and stored for off-line processing. This strong requirement must meet the need to observe muon events with each single telescope to allow for calibrations. In this scenario, the correct identification of the single telescope triggers participating to the same stereo event is of fundamental importance. In this contribution we present the SEB constraints, design, and expected performance.
- 25 July 2023
- Published by Sissa Medialab Srl
- Vol. 444, 597
Open AccessThe Stereo Event Builder software system of the ASTRI Mini-Array
The Stereo Event Builder (SEB) software system of the ASTRI Mini-Array is the part of the off-line data processing chain that is responsible for identifying single and stereo Cherenkov events. The ASTRI Mini-Array is an international project led by the Italian National Institute for Astrophysics (INAF) which is in the process of deploying nine Imaging Atmospheric Cherenkov Telescopes (IACTs) of the 4-m class at the Observatorio del Teide in Tenerife (Spain). The project is designed to detect very high-energy gamma rays up to the multi-TeV scale. In the ASTRI Mini-Array operation concept, the expected performance is based on the stereoscopic technique, i.e. the detection of the same atmospheric shower event with two or more telescopes. However all single-telescope events are acquired independently and stored for off-line processing. This strong requirement must meet the need to observe muon events with each single telescope to allow for calibrations. In this scenario, the correct identification of the single telescope triggers participating to the same stereo event is of fundamental importance. In this contribution we present the SEB constraints, design, and expected performance.
- Open Access
- 13 July 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (14) , 8172
Application of Machine Learning Ensemble Methods to ASTRI Mini-Array Cherenkov Event Reconstruction
The Imaging Atmospheric Cherenkov technique has opened up previously unexplored windows for the study of astrophysical radiation sources in the very high-energy (VHE) regime and is playing an important role in the discovery and characterization of VHE gamma-ray emitters. However, even for the most powerful sources, the data collected by Imaging Atmospheric Cherenkov Telescopes (IACTs) are heavily dominated by the overwhelming background due to cosmic-ray nuclei and cosmic-ray electrons. As a result, the analysis of IACT data necessitates the use of a highly efficient background rejection technique capable of distinguishing a gamma-ray induced signal through identification of shape features in its image. We present a detailed case study of gamma/hadron separation and energy reconstruction. Using a set of simulated data based on the ASTRI Mini-Array Cherenkov telescopes, we have assessed and compared a number of supervised Machine Learning methods, including the Random Forest method, Extra Trees method, and Extreme Gradient Boosting (XGB). To determine the optimal weighting for each method in the ensemble, we conducted extensive experiments involving multiple trials and cross-validation tests. As a result of this thorough investigation, we found that the most sensitive Machine Learning technique applied to our data sample for gamma/hadron segregation is a Stacking Ensemble Method composed of 42% Extra Trees, 28% Random Forest, and 30% XGB. In addition, the best-performing technique for energy estimation is a different Stacking Ensemble Method composed of 45% XGB, 27.5% Extra Trees, and 27.5% Random Forest. These optimal weightings were derived from extensive testing and fine-tuning, ensuring maximum performance for both gamma/hadron separation and energy estimation.
- 13 July 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (14) , 8172
Open AccessApplication of Machine Learning Ensemble Methods to ASTRI Mini-Array Cherenkov Event Reconstruction
The Imaging Atmospheric Cherenkov technique has opened up previously unexplored windows for the study of astrophysical radiation sources in the very high-energy (VHE) regime and is playing an important role in the discovery and characterization of VHE gamma-ray emitters. However, even for the most powerful sources, the data collected by Imaging Atmospheric Cherenkov Telescopes (IACTs) are heavily dominated by the overwhelming background due to cosmic-ray nuclei and cosmic-ray electrons. As a result, the analysis of IACT data necessitates the use of a highly efficient background rejection technique capable of distinguishing a gamma-ray induced signal through identification of shape features in its image. We present a detailed case study of gamma/hadron separation and energy reconstruction. Using a set of simulated data based on the ASTRI Mini-Array Cherenkov telescopes, we have assessed and compared a number of supervised Machine Learning methods, including the Random Forest method, Extra Trees method, and Extreme Gradient Boosting (XGB). To determine the optimal weighting for each method in the ensemble, we conducted extensive experiments involving multiple trials and cross-validation tests. As a result of this thorough investigation, we found that the most sensitive Machine Learning technique applied to our data sample for gamma/hadron segregation is a Stacking Ensemble Method composed of 42% Extra Trees, 28% Random Forest, and 30% XGB. In addition, the best-performing technique for energy estimation is a different Stacking Ensemble Method composed of 45% XGB, 27.5% Extra Trees, and 27.5% Random Forest. These optimal weightings were derived from extensive testing and fine-tuning, ensuring maximum performance for both gamma/hadron separation and energy estimation.
- Open Access
- 6 May 2022
- Published by Elsevier in Journal of High Energy Astrophysics
- Vol. 35, 52-68
The ASTRI Mini-Array of Cherenkov telescopes at the Observatorio del Teide
- 6 May 2022
- Published by Elsevier in Journal of High Energy Astrophysics
- Vol. 35, 52-68
Open AccessThe ASTRI Mini-Array of Cherenkov telescopes at the Observatorio del Teide