Pith. sign in

REVIEW 2 major objections 5 minor 48 references

Partitioning CTAO events by predicted direction error and analyzing them with type-specific response functions improves sensitivity by about 25% and spatial resolving power by 25-50%.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · grok-4.5

2026-07-13 04:07 UTC pith:A52EMSZD

load-bearing objection Solid end-to-end MC demonstration that event-type partitioning gives CTAO ~25% sensitivity and 25–50% angular-resolution gains; the only real soft spot is the untested transfer to real data. the 2 major comments →

arxiv 2607.09286 v1 pith:A52EMSZD submitted 2026-07-10 astro-ph.IM astro-ph.HE

Enhancing the Cherenkov Telescope Array Observatory high-level performance through an event-type-based analysis

Juan Bernete , Silvia Garc\'ia-Soto , Tarek Hassan , Orel Gueta , Maximilian Linhoff , Atreyee Sinha , Gernot Maier This is my paper
classification astro-ph.IM astro-ph.HE
keywords gamma-ray astronomyCherenkov telescopesCTAOInstrument response functionsEvent-type analysisMachine learningAngular resolutionSensitivity
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Standard Imaging Atmospheric Cherenkov Telescope analysis applies quality cuts, discards lower-quality events, and treats every surviving event as having the same average instrument response. This paper shows that the opposite strategy works better for the future Cherenkov Telescope Array Observatory. A multi-layer perceptron predicts each event's direction-reconstruction error (misdirection); events are then ranked and split into quality-based types. Separate instrument response functions are built for each type, and the types are analyzed jointly as independent observations. On Monte-Carlo simulations the combined analysis recovers events that would otherwise have been thrown away, raises differential sensitivity by roughly 25%, and improves angular resolution of the best events by 25-50%. The gain matters for crowded fields such as the Galactic Plane and for searches that need sharp spatial or spectral features.

Core claim

An event-type analysis that ranks CTAO simulated events by multi-layer-perceptron-predicted misdirection, builds type-specific instrument response functions, and analyzes the types jointly improves combined differential sensitivity by approximately 25% (performance-per-unit-time ~1.27 North / 1.22 South) and spatial resolving power by 25-50% relative to the conventional single-IRF analysis that discards lower-quality events.

What carries the argument

Misdirection-ranked event types: a multi-layer perceptron predicts the angular difference between true and reconstructed gamma-ray direction; thresholds on that continuous score partition the data into independent quality classes, each with its own instrument response functions that are later combined in a joint high-level fit.

Load-bearing premise

The Monte-Carlo simulations used to train the predictor and to compute the response functions faithfully reproduce the reconstruction-quality distributions and gamma-hadron separation that real CTAO telescopes will deliver.

What would settle it

Apply the identical misdirection-prediction and event-type pipeline to real data from the Large-Sized Telescope prototype (LST-1) and measure whether the combined sensitivity and angular-resolution gains remain at the levels reported on the simulations.

Watch this falsifier — get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. The paper presents a proof-of-concept event-type analysis for CTAO using Prod5v0.1 Monte Carlo simulations. An MLP is trained (in 20 energy quantile bins) to predict logarithmic direction reconstruction error (misdirection) from image and array parameters; events are then partitioned into four types with fixed gamma-ray fractions (15/15/30/40 %). Type-specific IRFs are produced with pyirf after independent cut optimization, and joint high-level analyses are performed with Gammapy. Relative to the standard single-IRF pipeline the authors report a combined differential sensitivity improvement of ~25 % (PPUT ≈ 1.27 North / 1.22 South) and a 25–50 % gain in spatial resolving power, demonstrated via extension-detection and close-source-separation likelihood-ratio tests.

Significance. If the gains survive the transition to real CTAO data they would be scientifically important for source-confused fields (Galactic Plane Survey), morphology studies, IGMF halo searches and spectral-line dark-matter analyses. The work is transparent: the MLP architecture, training/test split, random-partition control, multiple fraction configurations, and public tools (EventDisplay, pyirf, Gammapy) are all documented; the event-type reconstruction code is released. These elements make the result reproducible within the simulation framework and constitute a clear methodological advance over the conventional single-IRF IACT analysis.

major comments (2)
  1. [§2.5.1 / Fig. 3] §2.5.1 and Fig. 3: the joint sensitivity (and therefore the headline ~25 % PPUT) is obtained exclusively with the Forward-folding / Cash-statistic estimator. That estimator is shown to return fluxes ~25 % higher than the standard Li & Ma + N_excess≥10 + Bkg-fraction cuts at the highest energies, and it cannot enforce those cuts. While the relative comparison is internally consistent, the abstract and §3.1 statements of “~25 % in sensitivity” should be explicitly qualified as relative under this particular estimator; an additional joint Li & Ma-style calculation (or a hybrid) would strengthen the claim.
  2. [§3.1.2 / Table 1] §3.1.2 / Table 1 and §4: the chosen four-type partition already reduces the MC statistics available for the rarest event types; the paper notes that five types become statistics-limited but does not propagate the resulting IRF uncertainties into the PPUT, AP or likelihood-ratio significances. A simple bootstrap or jackknife estimate of the PPUT variance would quantify whether the reported 25 % gain remains significant once finite-MC and MLP-training fluctuations are included.
minor comments (5)
  1. [§4] §4, second paragraph: typographical duplication “These results were were calculated”.
  2. [Introduction] Introduction: “multiwavelenght” → “multiwavelength”.
  3. [Fig. 2] Fig. 2 caption and main text: the energy bin shown in the right panel is stated as 0.50–0.79 TeV; confirm consistency with the 20 quantile bins used for training.
  4. [Appendix A] Appendix A / Fig. A.12: the colour scale for Garson ranks is inverted relative to the usual “high-importance = dark” convention; a short note would avoid misreading.
  5. [Data Availability] Data-availability statement: the Zenodo DOI for the event-type code is given, but a short README describing the exact Prod5v0.1 subset and EventDisplay version used would improve long-term reproducibility.

Circularity Check

0 steps flagged

No significant circularity: gains are measured empirically on held-out MC after independent quality partitioning, not forced by definition or self-citation.

full rationale

The paper's chain is: (1) train an MLP regressor on 25% of Prod5 MC to predict log-misdirection from image/array features (Sec. 2.3, Fig. 2); (2) apply energy/offset-binned thresholds on the predicted quantity to the held-out 75% so that gamma-ray fractions match chosen partitions (optimized via PPUT scan in Sec. 3.1.2); (3) compute independent type-wise IRFs with pyirf (Sec. 2.4, Figs. 7-8); (4) feed those IRFs into Gammapy joint analyses and measure sensitivity (forward-folding), angular performance, extension significance and source-separation significance against the single-IRF baseline (Secs. 2.5, 3.1-3.3). None of these steps reduces the reported ~25% PPUT or 25-50% resolving-power numbers to the training labels or to a fitted parameter by construction. A random-partition control (Sec. 3.1.2) yields PPUT consistent with 1, confirming the gain is not an artifact of the MC sample size or of the IRF machinery. Self-citations ([13],[14]) are earlier conference/arXiv versions of the same idea and are not load-bearing uniqueness theorems; the method is motivated by the external Fermi-LAT precedent. Using the same MC production for training and IRF evaluation is standard practice and does not embed the final performance metrics into the event-type definition. The work is therefore a self-contained empirical demonstration on simulations, not a circular derivation.

Axiom & Free-Parameter Ledger

4 free parameters · 3 axioms · 0 invented entities

The central performance claims rest on standard IACT Monte-Carlo practice, a modest set of free architectural and partitioning choices, and the domain assumption that simulated reconstruction quality maps to real data. No new physical entities are postulated.

free parameters (4)
  • MLP architecture (2 hidden layers 36/6, tanh, max 2e4 iterations, tol 1e-5)
    Chosen after informal comparison with random forests and SVRs; hyperparameters not systematically optimized or cross-validated beyond the reported energy-binned training.
  • Event-type gamma-ray fractions (15%/15%/30%/40%)
    Selected after a discrete scan of 2–5 partitions by maximizing mean PPUT; the exact fractions are free choices that affect the reported gains.
  • Number of energy bins for MLP training (20 quantile bins)
    Ad-hoc choice to capture energy dependence; not derived from first principles.
  • Quality-cut optimization criteria (5σ Li&Ma, ≥10 excess, ≥5% background)
    Standard CTAO sensitivity definition, but still a free analysis choice that influences which events survive into each IRF.
axioms (3)
  • domain assumption CORSIKA + sim_telarray + EventDisplay Prod5v0.1 simulations accurately model CTAO telescope response, shower development, and reconstruction biases for both gamma rays and hadronic background.
    Invoked throughout §§2.1–2.4 and for all IRF and high-level results; the paper itself flags that real-data validation is still required.
  • ad hoc to paper Predicted logarithmic misdirection is a sufficient scalar proxy for overall reconstruction quality (direction and energy).
    Used to define event types (§2.3); energy resolution improves even though it was never an explicit training target, but the proxy is not proven optimal.
  • domain assumption Joint likelihood analysis of independent event-type datasets with Gammapy correctly combines information without introducing bias from the correlation between angular and energy resolution.
    Stated in §2.5 and §4; the paper notes that full multi-dimensional IRFs would be needed to treat the correlation rigorously.

pith-pipeline@v1.1.0-grok45 · 24188 in / 2869 out tokens · 27368 ms · 2026-07-13T04:07:26.207337+00:00 · methodology

0 comments
read the original abstract

The analysis traditionally employed by Imaging Atmospheric Cherenkov Telescopes involves optimizing quality cuts to select a sub-sample of high-quality events. These events are used for the scientific interpretation of the data, employing a single set of Instrument Response Functions (IRFs). All selected events are treated equally and assumed to be well represented by these IRFs, while the rest are discarded. An alternative approach, successfully applied in experiments such as Fermi-LAT, is an event-type-based analysis. This method divides datasets into subsamples, each containing events of a given expected reconstruction quality. IRFs are computed for each subsample independently, improving the accuracy with which IRFs represent the reconstruction quality of each event. The high-level analysis of these subsamples is performed treating them as independent observations, each with their own set of IRFs, and analyzed jointly. In this work we present a proof-of-concept implementation of an event-type-based analysis for the future CTAO using simulated data. A neural network (specifically a multi-layer perceptron) is trained to predict the direction reconstruction error of each event, and the simulated dataset is divided into event types based on this predicted variable. We compute IRFs for each event type and compare them with those from the standard analysis (without event types). Finally, we simulate observations using these event-type-wise IRFs and analyze them with high-level analysis tools to test the performance of both approaches. This implementation demonstrates notable improvements: 25% to 50% boost in spatial resolving power and ~25% in sensitivity. This boost in performance will have strong implications in the scientific exploitation of the CTAO data, especially in crowded regions such as the Galactic Plane or searching for spectral signatures like Dark Matter annihilation lines.

Figures

Figures reproduced from arXiv: 2607.09286 by Atreyee Sinha, Gernot Maier, Juan Bernete, Maximilian Linhoff, Orel Gueta, Silvia Garc\'ia-Soto, Tarek Hassan.

Figure 1
Figure 1. Figure 1: Diagram outlining the steps of the methodology used in this work. ET is used [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Left) Predicted logarithm of the misdirection for gamma-ray events as a function of reconstructed energy. The assigned event type is shown in different colors. Middle) True logarithm of the misdirection for gamma-ray events as a function of reconstructed energy. Colors represent the assigned event types. Solid lines indicate the median of the distribution of each event type. Right) Predicted vs true logari… view at source ↗
Figure 3
Figure 3. Figure 3: Comparison of the differential sensitivity computed with the 3 different methods [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Point-source sensitivity for both arrays at offset 0.5 degrees from the pointing [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: PPUTs for both arrays showing the general improvement of the combined sen [PITH_FULL_IMAGE:figures/full_fig_p018_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: PPUTs for both arrays showing the general improvement of the combined sen [PITH_FULL_IMAGE:figures/full_fig_p019_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Effective area, background rate, angular resolution and energy resolution for [PITH_FULL_IMAGE:figures/full_fig_p021_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Effective area, background rate, angular resolution and energy resolution for [PITH_FULL_IMAGE:figures/full_fig_p022_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Angular Performances (APs) for the angular resolution of an event type 1 [PITH_FULL_IMAGE:figures/full_fig_p024_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Significance of the detection of the source extension for different radii and [PITH_FULL_IMAGE:figures/full_fig_p025_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Significance of the detection two separated sources for different separations [PITH_FULL_IMAGE:figures/full_fig_p026_11.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

48 extracted references · 17 canonical work pages · 1 internal anchor

  1. [1]

    Zanin, et al., CTA – the World’s largest ground-based gamma-ray observatory, PoS ICRC2021 (2021) 005

    R. Zanin, et al., CTA – the World’s largest ground-based gamma-ray observatory, PoS ICRC2021 (2021) 005. doi:10.22323/1.395.0005

    work page doi:10.22323/1.395.0005 2021

  2. [2]

    doi:10.1142/10986

    Cherenkov Telescope Array Consortium, et al., Science with the Cherenkov Telescope Array, World Scientific, 2019. doi:10.1142/10986. 31 Figure A.12: Grid showing the ranking of importance calculated using Garson’s algorithm. Annotations in the grid represent the rank of that particular variable and energy bin, with the x axis ordered by the overall import...

    doi:10.1142/10986 2019

  3. [3]

    5281/zenodo.5499840

    Cherenkov Telescope Array Observatory and Cherenkov Telescope Ar- ray Consortium, Ctao instrument response functions - prod5 version v0.1, 2021.URL:https://doi.org/10.5281/zenodo.5499840.doi:10. 5281/zenodo.5499840

    work page doi:10.5281/zenodo.5499840.doi:10 2021

  4. [4]

    Bernlöhr, O

    K. Bernlöhr, O. Gueta, G. Maier, A. Moralejo, Y. Suda, CTAO Simulation Telescope Models for CORSIKA and sim_telarray - prod5, 2022. URL:https://doi.org/10.5281/zenodo.6218687. doi:10.5281/zenodo.6218687

    work page doi:10.5281/zenodo.6218687 2022

  5. [5]

    Bernlöhr, et al., Monte carlo design studies for the cherenkov tele- scope array, Astroparticle Physics 43 (2013) 171–188

    K. Bernlöhr, et al., Monte carlo design studies for the cherenkov tele- scope array, Astroparticle Physics 43 (2013) 171–188. URL:https: //doi.org/10.1016/j.astropartphys.2012.10.002. doi:10.1016/j. astropartphys.2012.10.002

    doi:10.1016/j.astropartphys.2012.10.002 2013

  6. [6]

    Hassan, et al., Monte carlo performance studies for the site selection of the cherenkov telescope array, Astroparticle Physics 93 (2017) 76–85

    T. Hassan, et al., Monte carlo performance studies for the site selection of the cherenkov telescope array, Astroparticle Physics 93 (2017) 76–85. doi:https://doi.org/10.1016/j.astropartphys.2017.05.001. 32

    work page doi:10.1016/j.astropartphys.2017.05.001 2017

  7. [7]

    Acharyya, et al., Monte carlo studies for the optimisation of the cherenkov telescope array layout, Astroparticle Physics 111 (2019) 35–53

    A. Acharyya, et al., Monte carlo studies for the optimisation of the cherenkov telescope array layout, Astroparticle Physics 111 (2019) 35–53. URL:https://doi.org/10.1016/j.astropartphys.2019.04

    work page doi:10.1016/j.astropartphys.2019.04 2019

  8. [8]

    doi:10.1016/j.astropartphys.2019.04.001

    work page doi:10.1016/j.astropartphys.2019.04.001 2019

  9. [9]

    Donath, et al., Gammapy: A python package for gamma-ray as- tronomy, A&A 678 (2023) A157

    A. Donath, et al., Gammapy: A python package for gamma-ray as- tronomy, A&A 678 (2023) A157. URL:https://doi.org/10.1051/ 0004-6361/202346488. doi:10.1051/0004-6361/202346488

    doi:10.1051/0004-6361/202346488 2023

  10. [10]

    Knödlseder, et al., GammaLib and ctools

    J. Knödlseder, et al., GammaLib and ctools. A software frame- work for the analysis of astronomical gamma-ray data, Astronomy & Astrophysics 593 (2016) A1. doi:10.1051/0004-6361/201628822. arXiv:1606.00393

    Pith/arXiv arXiv doi:10.1051/0004-6361/201628822 2016

  11. [11]

    D’Amico, T

    G. D’Amico, T. Terzić, J. Strišković, M. Doro, M. Strzys, J. van Scherpenberg, Signal estimation in on/off measurements including event-by-event variables, Phys. Rev. D 103 (2021) 123001. URL: https://link.aps.org/doi/10.1103/PhysRevD.103.123001. doi:10. 1103/PhysRevD.103.123001

    work page doi:10.1103/physrevd.103.123001 2021

  12. [12]

    W. B. Atwood, A. A. Abdo, et al., The large area telescope on the fermi gamma-ray space telescope mission, The Astrophysical Journal 697 (2009) 1071. URL:https://dx.doi.org/10.1088/0004-637X/697/2/

    doi:10.1088/0004-637x/697/2/ 2009

  13. [13]

    doi:10.1088/0004-637X/697/2/1071

    doi:10.1088/0004-637x/697/2/1071

  14. [14]

    Atwood, et al., Pass 8: Toward the full realization of the fermi-lat scientific potential, 2013

    W. Atwood, et al., Pass 8: Toward the full realization of the fermi-lat scientific potential, 2013. URL:https://arxiv.org/abs/1303.3514. arXiv:1303.3514

    Pith/arXiv arXiv 2013

  15. [15]

    Hassan, O

    T. Hassan, O. Gueta, G. Maier, M. Nöthe, M. Peresano, I. Vovk, Per- formance of a proposed event-type based analysis for CTA, in: Proceed- ings of 37th International Cosmic Ray Conference — PoS(ICRC2021), Sissa Medialab, 2021, p. 752. URL:https://doi.org/10.22323/1. 395.0752. doi:10.22323/1.395.0752

    doi:10.22323/1 2021

  16. [16]

    Bernete, O

    J. Bernete, O. Gueta, T. Hassan, M. Linhoff, G. Maier, A. Sinha, Perfor- mance update of an event-type based analysis for the Cherenkov Tele- scope Array, arXiv e-prints (2023) arXiv:2309.11375. doi:10.48550/ arXiv.2309.11375.arXiv:2309.11375. 33

    Pith/arXiv arXiv 2023

  17. [17]

    Abdalla, et al., The h.e.s.s

    H. Abdalla, et al., The h.e.s.s. galactic plane survey, Astronomy & Astrophysics 612 (2018) A1. URL:http://dx.doi.org/10.1051/ 0004-6361/201732098. doi:10.1051/0004-6361/201732098

    doi:10.1051/0004-6361/201732098 2018

  18. [18]

    H. Abdalla, et al., Sensitivity of the cherenkov telescope array for probing cosmology and fundamental physics with gamma-ray propa- gation, Journal of Cosmology and Astroparticle Physics 2021 (2021) 048–048. URL:https://doi.org/10.1088/1475-7516%2F2021%2F02% 2F048. doi:10.1088/1475-7516/2021/02/048

    doi:10.1088/1475-7516 2021

  19. [19]

    Acharyya, et al., Sensitivity of the cherenkov telescope array to a dark matter signal from the galactic centre, Journal of Cosmology and Astroparticle Physics 2021 (2021) 057

    A. Acharyya, et al., Sensitivity of the cherenkov telescope array to a dark matter signal from the galactic centre, Journal of Cosmology and Astroparticle Physics 2021 (2021) 057. URL:https://doi.org/ 10.1088/1475-7516/2021/01/057. doi:10.1088/1475-7516/2021/01/ 057

    doi:10.1088/1475-7516/2021/01/057 2021

  20. [20]

    K. Abe, et al., Prospects for dark matter observations in dwarf spheroidal galaxies with the cherenkov telescope array observa- tory, Monthly Notices of the Royal Astronomical Society 544 (2025) 2946–2986. URL:https://doi.org/10.1093/mnras/staf1798. doi:10.1093/mnras/staf1798

    work page doi:10.1093/mnras/staf1798 2025

  21. [21]

    A. Acharyya, et al., Sensitivity of the cherenkov telescope array to tev photon emission from the large magellanic cloud, Monthly Notices of the Royal Astronomical Society 523 (2023) 5353–5387. URL:https: //doi.org/10.1093/mnras/stad1576. doi:10.1093/mnras/stad1576

    doi:10.1093/mnras/stad1576 2023

  22. [22]

    Abe, et al., Dark matter line searches with the cherenkov telescope array, Journal of Cosmology and Astroparticle Physics 2024 (2024) 047

    S. Abe, et al., Dark matter line searches with the cherenkov telescope array, Journal of Cosmology and Astroparticle Physics 2024 (2024) 047. URL:http://dx.doi.org/10.1088/1475-7516/2024/07/047. doi:10. 1088/1475-7516/2024/07/047

    doi:10.1088/1475-7516/2024/07/047 2024

  23. [23]

    D. Heck, J. Knapp, J. N. Capdevielle, G. Schatz, T. Thouw, COR- SIKA: a Monte Carlo code to simulate extensive air showers., Tech- nical Report FZKA 6019, Forschungszentrum Karlsruhe, 1998. URL: https://publikationen.bibliothek.kit.edu/270043064/3813660

    arXiv 1998

  24. [24]

    Bernlöhr, Simulation of imaging atmospheric Cherenkov telescopes with CORSIKA and sim_ telarray, Astroparticle Physics 30 (2008) 149–

    K. Bernlöhr, Simulation of imaging atmospheric Cherenkov telescopes with CORSIKA and sim_ telarray, Astroparticle Physics 30 (2008) 149–

    2008

  25. [25]

    doi:10.1016/j.astropartphys.2008.07.009.arXiv:0808.2253. 34

    Pith/arXiv arXiv doi:10.1016/j.astropartphys.2008.07.009.arxiv:0808.2253 2008

  26. [26]

    Ostapchenko, Monte carlo treatment of hadronic interac- tions in enhanced pomeron scheme: Qgsjet-ii model, Phys

    S. Ostapchenko, Monte carlo treatment of hadronic interac- tions in enhanced pomeron scheme: Qgsjet-ii model, Phys. Rev. D 83 (2011) 014018. URL:https://link.aps.org/doi/10.1103/ PhysRevD.83.014018. doi:10.1103/PhysRevD.83.014018

    doi:10.1103/physrevd.83.014018 2011

  27. [27]

    Bass, Microscopic models for ultrarelativistic heavy ion collisions, Progress in Particle and Nuclear Physics 41 (1998) 255–369

    S. Bass, Microscopic models for ultrarelativistic heavy ion collisions, Progress in Particle and Nuclear Physics 41 (1998) 255–369. URL: http://dx.doi.org/10.1016/S0146-6410(98)00058-1. doi:10.1016/ s0146-6410(98)00058-1

    doi:10.1016/s0146-6410(98)00058-1 1998

  28. [28]

    M. Bleicher, et al., Relativistic hadron-hadron collisions in the ultra- relativistic quantum molecular dynamics model, Journal of Physics G: Nuclear and Particle Physics 25 (1999) 1859. URL:https://dx. doi.org/10.1088/0954-3899/25/9/308. doi:10.1088/0954-3899/25/ 9/308

    doi:10.1088/0954-3899/25/9/308 1999

  29. [29]

    Maier, J

    G. Maier, J. Holder, Eventdisplay: An Analysis and Reconstruction Package for Ground-based Gamma-ray Astronomy, in: Proceedings, 35th International Cosmic Ray Conference, volume 301, 2017, p. 747. doi:10.22323/1.301.0747

    doi:10.22323/1.301.0747 2017

  30. [30]

    Maier, et al., Eventdisplay: An Analysis and Reconstruction Package for Ground-based Gamma-ray Astronomy, 2024

    G. Maier, et al., Eventdisplay: An Analysis and Reconstruction Package for Ground-based Gamma-ray Astronomy, 2024. doi:10.5281/zenodo. 11096726

    doi:10.5281/zenodo 2024

  31. [31]

    Pedregosa, et al., Scikit-learn: Machine learning in Python, Journal of Machine Learning Research 12 (2011) 2825–2830

    F. Pedregosa, et al., Scikit-learn: Machine learning in Python, Journal of Machine Learning Research 12 (2011) 2825–2830

    2011

  32. [32]

    G. D. Garson, Interpreting neural-network connection weights, Artificial Intelligence Expert 6 (1991) 46–51

    1991

  33. [33]

    Goh, Back-propagation neural networks for modeling complex sys- tems, Artificial Intelligence in Engineering 9 (1995) 143–151

    A. Goh, Back-propagation neural networks for modeling complex sys- tems, Artificial Intelligence in Engineering 9 (1995) 143–151. doi:https: //doi.org/10.1016/0954-1810(94)00011-S

    work page doi:10.1016/0954-1810(94)00011-s 1995

  34. [34]

    Schwefer, R

    G. Schwefer, R. Parsons, J. Hinton, A hybrid approach to event re- construction for atmospheric cherenkov telescopes combining machine learning and likelihood fitting, 2024. URL:https://arxiv.org/abs/ 2406.17502.arXiv:2406.17502. 35

    Pith/arXiv arXiv 2024

  35. [35]

    T. P. Li, Y. Q. Ma, Analysis methods for results in gamma-ray as- tronomy., The Astrophysical Journal 272 (1983) 317–324. doi:10.1086/ 161295

    1983

  36. [36]

    URL:https:// www.mdpi.com/2218-1997/7/10/374

    C.Nigro, T.Hassan, L.Olivera-Nieto, Evolutionofdataformatsinvery- high-energy gamma-ray astronomy, Universe 7 (2021). URL:https:// www.mdpi.com/2218-1997/7/10/374. doi:10.3390/universe7100374

    work page doi:10.3390/universe7100374 2021

  37. [37]

    Deil, et al., Open high-level data formats and software for gamma- ray astronomy, in: AIP Conference Proceedings, volume 1792, AIP Publishing LLC, 2017, p

    C. Deil, et al., Open high-level data formats and software for gamma- ray astronomy, in: AIP Conference Proceedings, volume 1792, AIP Publishing LLC, 2017, p. 070006

    2017

  38. [38]

    Linhoff, et al., pyirf, 2025

    M. Linhoff, et al., pyirf, 2025. URL:https://github.com/ cta-observatory/pyirf. doi:10.5281/zenodo.15488321

    work page doi:10.5281/zenodo.15488321 2025

  39. [39]

    Aguasca-Cabot, et al., Gammapy: Python toolbox for gamma-ray astronomy, 2023

    A. Aguasca-Cabot, et al., Gammapy: Python toolbox for gamma-ray astronomy, 2023. URL:https://doi.org/10.5281/zenodo.8033275. doi:10.5281/zenodo.8033275

    work page doi:10.5281/zenodo.8033275 2023

  40. [40]

    Acero, et al., Gammapy: Python toolbox for gamma-ray as- tronomy, 2025

    F. Acero, et al., Gammapy: Python toolbox for gamma-ray as- tronomy, 2025. URL:https://doi.org/10.5281/zenodo.17814297. doi:10.5281/zenodo.17814297

    work page doi:10.5281/zenodo.17814297 2025

  41. [41]

    Cash, Parameter estimation in astronomy through application of the likelihood ratio., The Astrophysical Journal 228 (1979) 939–947

    W. Cash, Parameter estimation in astronomy through application of the likelihood ratio., The Astrophysical Journal 228 (1979) 939–947. doi:10.1086/156922

    doi:10.1086/156922 1979

  42. [42]

    Aharonian, et al., The crab nebula and pulsar between 500 gev and 80 tev: Observations with the hegra stereoscopic air cerenkov telescopes, The Astrophysical Journal 614 (2004) 897

    F. Aharonian, et al., The crab nebula and pulsar between 500 gev and 80 tev: Observations with the hegra stereoscopic air cerenkov telescopes, The Astrophysical Journal 614 (2004) 897. URL:https://dx.doi.org/ 10.1086/423931. doi:10.1086/423931

    work page doi:10.1086/423931 2004

  43. [43]

    Abe, et al., Prospects for a survey of the galactic plane with the cherenkov telescope array, Journal of Cosmology and Astropar- ticle Physics 2024 (2024) 081

    S. Abe, et al., Prospects for a survey of the galactic plane with the cherenkov telescope array, Journal of Cosmology and Astropar- ticle Physics 2024 (2024) 081. URL:https://dx.doi.org/10.1088/ 1475-7516/2024/10/081. doi:10.1088/1475-7516/2024/10/081

    work page doi:10.1088/1475-7516/2024/10/081 2024

  44. [44]

    Zanin, et al., MARS, The MAGIC Analysis and Reconstruction Software, in: International Cosmic Ray Conference, volume 33 ofInter- national Cosmic Ray Conference, 2013, p

    R. Zanin, et al., MARS, The MAGIC Analysis and Reconstruction Software, in: International Cosmic Ray Conference, volume 33 ofInter- national Cosmic Ray Conference, 2013, p. 2937. 36

    2013

  45. [45]

    D. Morcuende, et al., Performance of the large-sized telescope proto- type of the cherenkov telescope array, in: Proceedings of 38th Interna- tional Cosmic Ray Conference — PoS(ICRC2023), ICRC2023, Sissa Me- dialab, 2023, p.594.URL:http://dx.doi.org/10.22323/1.444.0594. doi:10.22323/1.444.0594

    work page doi:10.22323/1.444.0594 2023

  46. [46]

    Bernete, O

    J. Bernete, O. Gueta, T. Hassan, Iact event types, 2025. URL:https: //github.com/cta-observatory/iact_event_types

    2025

  47. [47]

    A. M. Hillas, Cerenkov Light Images of EAS Produced by Primary Gamma Rays and by Nuclei, in: F. C. Jones (Ed.), 19th International Cosmic Ray Conference (ICRC19), Volume 3, volume 3 ofInternational Cosmic Ray Conference, 1985, p. 445

    1985

  48. [48]

    Gamma-Hadron Separation Methods for the VERITAS Array of Four Imaging Atmospheric Cherenkov Telescopes

    H. Krawczynski, et al., Gamma hadron separation methods for the VERITAS array of four imaging atmospheric Cherenkov telescopes, As- troparticle Physics 25 (2006) 380–390. doi:10.1016/j.astropartphys. 2006.03.011.arXiv:astro-ph/0604508. 37

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1016/j.astropartphys 2006