pith. sign in

arxiv: 2606.09216 · v1 · pith:X7RRFK7Nnew · submitted 2026-06-08 · ⚛️ physics.ins-det

An investigation of fast simulation techniques for pion showers using kernel density estimators with the CALICE AHCAL Technological Prototype

CALICE Collaboration: A. Wilhahn , J. Utehs , Z. Ghafoor , G. Eigen , S. Lai , O. Bach , E. Brianne , K. Gadow
show 23 more authors
D. Heuchel K. Kr\"uger J. Kvasnicka A. Laudrain O. Pinto M. Reinecke F. Sefkow M. De Silva E. Garutti G. Kasieczka S. Martens J. Rolph F. Hummer F. Simon A. Brogna V. B\"uscher L. Masetti A. Rosmanitz C. Schmitt Q. Weitzel W. Ootani T. Suehara A. Irles
This is my paper

Pith reviewed 2026-06-27 14:29 UTC · model grok-4.3

classification ⚛️ physics.ins-det
keywords fast simulationkernel density estimatorspion showersAHCALCALICEhadron calorimetrytest beam
0
0 comments X

The pith

A kernel density estimator trained on 2018 pion test beam data reproduces AHCAL shower observables and supports interpolation across energies.

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

The paper develops a data-driven fast simulation for the response of the CALICE AHCAL Technological Prototype to pion showers. It trains kernel density estimators directly on measured shower properties from negatively charged pions between 10 and 200 GeV recorded at CERN in 2018. The resulting model matches measured shower observables closely. An interpolation procedure then combines simulations at neighboring beam energies to produce showers at any intermediate energy.

Core claim

The kernel density estimator model built from the 2018 test beam dataset produces synthetic pion showers whose observables agree closely with the measured data in the AHCAL prototype, and the interpolation method between neighboring energies extends the simulation capability to arbitrary energies within the covered range.

What carries the argument

Kernel density estimation applied to the distribution of shower observables extracted from the test beam events, used to sample new showers that follow the same statistical properties.

If this is right

  • The fast simulation can be used in place of full Monte Carlo for large-scale studies of detector response to pions.
  • Interpolation allows continuous coverage of energies between the discrete beam settings without additional full simulations.
  • The data-driven approach bypasses the need to tune hadronic interaction models for this specific calorimeter.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same kernel density technique could be tested on electron or muon data from the same run to check consistency across particle types.
  • If the interpolation holds, it might reduce the required number of dedicated beam energies in future test campaigns.
  • The method's success depends on the dimensionality of the observable space remaining manageable; higher-dimensional observables could require regularization not discussed here.

Load-bearing premise

The 2018 test beam dataset for pions between 10 and 200 GeV is representative enough that the kernel density estimator generalizes accurately across the full energy range and for the AHCAL prototype geometry.

What would settle it

A direct comparison showing statistically significant mismatch between simulated and measured shower observables, such as longitudinal or transverse profiles, at an energy either inside or outside the original beam energies.

Figures

Figures reproduced from arXiv: 2606.09216 by A. Brogna, A. Irles, A. Laudrain, A. Rosmanitz, CALICE Collaboration: A. Wilhahn, C. Schmitt, D. Heuchel, E. Brianne, E. Garutti, F. Hummer, F. Sefkow, F. Simon, G. Eigen, G. Kasieczka, J. Kvasnicka, J. Rolph, J. Utehs, K. Gadow, K. Kr\"uger, L. Masetti, M. De Silva, M. Reinecke, O. Bach, O. Pinto, Q. Weitzel, S. Lai, S. Martens, T. Suehara, V. B\"uscher, W. Ootani, Z. Ghafoor.

Figure 1
Figure 1. Figure 1: An example of a PDF estimation using Gaussian kernels. The hit energy [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Distribution of global CoGs in the xy-plane in 2018 Ahcal test beam data for 60 GeV pions. The x- and y-axis represent the position of the CoG and the colourbar depicts the normalised number of events. The majority of CoGs falls into a small region around the detector centre. Based on these observations, only events whose lateral CoGs fall within the range 12 [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Distributions of different kinematic shower variables for 60 GeV pions. [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Distributions of different shower moments for 60 GeV pions. The upper [PITH_FULL_IMAGE:figures/full_fig_p016_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Distributions of the same kinematic shower variables that have already been [PITH_FULL_IMAGE:figures/full_fig_p017_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Distributions of the same shower moments that have already been shown [PITH_FULL_IMAGE:figures/full_fig_p018_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Distributions for the number of hits per event for 60 GeV (left) and 120 GeV [PITH_FULL_IMAGE:figures/full_fig_p020_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Two-dimensional correlation plots for 60 GeV pion showers between the [PITH_FULL_IMAGE:figures/full_fig_p021_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Two-dimensional correlation plots for 120 GeV pion showers between the [PITH_FULL_IMAGE:figures/full_fig_p022_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Correlation matrices for kinematic shower variables obtained from data [PITH_FULL_IMAGE:figures/full_fig_p023_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Correlation matrices for kinematic shower variables obtained from data [PITH_FULL_IMAGE:figures/full_fig_p024_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Three-dimensional depictions of 60 GeV pion showers obtained from data [PITH_FULL_IMAGE:figures/full_fig_p025_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Illustrative example of integrating hit energy distributions. The upper [PITH_FULL_IMAGE:figures/full_fig_p028_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Distributions of different kinematic shower variables for 60 GeV pions. [PITH_FULL_IMAGE:figures/full_fig_p030_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: Distributions of different shower moments for 60 GeV pions. The upper [PITH_FULL_IMAGE:figures/full_fig_p031_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: Distributions of different kinematic shower variables for 120 GeV pions. [PITH_FULL_IMAGE:figures/full_fig_p032_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Distributions of different shower moments for 120 GeV pions. The mo [PITH_FULL_IMAGE:figures/full_fig_p033_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: Distributions for the number of hits per event for 60 GeV (left) and [PITH_FULL_IMAGE:figures/full_fig_p034_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: Two-dimensional correlation plots for 60 GeV pion showers between the [PITH_FULL_IMAGE:figures/full_fig_p035_19.png] view at source ↗
Figure 20
Figure 20. Figure 20: Two-dimensional correlation plots for 120 GeV pion showers between the [PITH_FULL_IMAGE:figures/full_fig_p036_20.png] view at source ↗
Figure 21
Figure 21. Figure 21: Correlation matrices for kinematic shower variables obtained from data [PITH_FULL_IMAGE:figures/full_fig_p037_21.png] view at source ↗
Figure 22
Figure 22. Figure 22: Correlation matrices for kinematic shower variables obtained from data [PITH_FULL_IMAGE:figures/full_fig_p038_22.png] view at source ↗
read the original abstract

In this article, the development and investigation of fast hadron shower simulation methods is presented. A test beam dataset has been recorded in 2018 at CERN with the AHCAL Technological Prototype of the CALICE Collaboration, where the calorimeter prototype was exposed to electron, muon, and negatively charged pion beams of various initial energies. The pion shower dataset, covering energies between 10 GeV and 200 GeV, has been used to develop a data-driven fast simulation algorithm of the AHCAL response to pion showers. The resulting shower model demonstrates excellent agreement with measured shower observables. In addition, a method for simulating pion showers at arbitrary energies is introduced, based upon interpolation between simulated showers at neighbouring beam energies.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript develops a data-driven fast simulation for pion showers in the CALICE AHCAL Technological Prototype using kernel density estimators trained on 2018 CERN test-beam data for negatively charged pions (10–200 GeV). It claims excellent agreement between the resulting model and measured shower observables and introduces an interpolation procedure to generate showers at arbitrary energies between the discrete training points.

Significance. If the quantitative validation holds, the approach could supply a computationally lightweight alternative to full GEANT4 hadronic simulations for calorimeter prototype studies, with the data-driven construction reducing reliance on hadronic interaction models.

major comments (2)
  1. [Abstract] Abstract: the assertion that 'the resulting shower model demonstrates excellent agreement with measured shower observables' is unsupported by any numerical metrics (e.g., χ^{2}, pull distributions, relative deviations, or Kolmogorov-Smirnov statistics), rendering the central claim of model fidelity impossible to assess from the supplied information.
  2. [Interpolation method] Interpolation method: the procedure for simulating showers at arbitrary energies by interpolating KDE models trained at neighbouring discrete beam energies implicitly assumes that shower observables (lateral/longitudinal profiles, hit multiplicity) vary sufficiently smoothly in KDE space. No hold-out validation at intermediate energies or test of non-linear energy dependence is described, which is load-bearing for the claim that the method works across the full 10–200 GeV range.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major point below and indicate the corresponding revisions.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the assertion that 'the resulting shower model demonstrates excellent agreement with measured shower observables' is unsupported by any numerical metrics (e.g., χ^{2}, pull distributions, relative deviations, or Kolmogorov-Smirnov statistics), rendering the central claim of model fidelity impossible to assess from the supplied information.

    Authors: We agree that the abstract would be strengthened by explicit quantitative metrics. While the manuscript body contains visual comparisons of distributions (energy sum, hit multiplicity, longitudinal and lateral profiles), we will revise the abstract to reference specific measures such as average relative deviations (typically <5% for integrated observables) and Kolmogorov-Smirnov statistics for profile shapes. This change will be made in the revised version. revision: yes

  2. Referee: [Interpolation method] Interpolation method: the procedure for simulating showers at arbitrary energies by interpolating KDE models trained at neighbouring discrete beam energies implicitly assumes that shower observables (lateral/longitudinal profiles, hit multiplicity) vary sufficiently smoothly in KDE space. No hold-out validation at intermediate energies or test of non-linear energy dependence is described, which is load-bearing for the claim that the method works across the full 10–200 GeV range.

    Authors: The interpolation procedure is grounded in the expectation of smooth energy dependence of hadronic shower observables, which is supported by the underlying physics and the discrete training points. Although a dedicated hold-out test at an intermediate energy was not presented in the original manuscript, consistency across the trained energies was verified. We will add a validation subsection that includes a hold-out comparison at an interpolated energy (e.g., using data at a point between training energies) and explicit checks for non-linear behaviour in key observables to substantiate the method over 10–200 GeV. revision: yes

Circularity Check

0 steps flagged

No circularity: model built directly from external test-beam data with no self-referential reductions

full rationale

The paper constructs its KDE-based pion shower model from the 2018 CERN test-beam dataset (external measurements on the AHCAL prototype) and reports agreement with the same measured observables. This is standard empirical fitting rather than any derivation that reduces to its inputs by construction. The interpolation method for arbitrary energies is an explicit post-processing step between discrete trained models and does not rename fitted parameters as predictions or invoke self-citations, uniqueness theorems, or ansatzes. No load-bearing step matches any of the enumerated circularity patterns; the chain remains self-contained against the external benchmark data.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review provides no information on free parameters, axioms, or invented entities.

pith-pipeline@v0.9.1-grok · 5810 in / 905 out tokens · 17759 ms · 2026-06-27T14:29:05.573019+00:00 · methodology

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

23 extracted references · 4 canonical work pages · 1 internal anchor

  1. [1]

    Agostinelli et al.Geant4—a simulation toolkit

    S. Agostinelli et al.Geant4—a simulation toolkit. Nucl. Instrum. Methods Phys. Res. A, 506:250–303, 2003

    2003

  2. [2]

    Allison et al.Geant4 developments and applications

    J. Allison et al.Geant4 developments and applications. IEEE Transactions on Nuclear Science, 53:270–278, 2006

    2006

  3. [3]

    Allison et al.Recent developments in Geant4

    J. Allison et al.Recent developments in Geant4. Nucl. Instrum. Methods Phys. Res. A, 835:186–225, 2016

    2016

  4. [4]

    ATLAS Collaboration.AtlFast3: The Next Generation of Fast Simulation in ATLAS. Comput. Softw. Big Sci., 6, 2022

    2022

  5. [5]

    CMS Collaboration.The Fast Simulation of The CMS Experiment. J. Phys. Conf. Ser., 396:062016, 2012

    2012

  6. [6]

    Rosenblatt.Remarks on Some Nonparametric Estimates of a Density Func- tion

    M. Rosenblatt.Remarks on Some Nonparametric Estimates of a Density Func- tion. Ann. Math. Statist., 27:832–837, 1956

    1956

  7. [7]

    Parzen.On Estimation of a Probability Density Function and Mode

    E. Parzen.On Estimation of a Probability Density Function and Mode. Ann. Math. Statist., 33:1065–1076, 1962

    1962

  8. [8]

    JINST, 18:P11018, 2022

    CALICE Collaboration.Design, Construction and Commissioning of a Techno- logical Prototype of a Highly Granular SiPM-on-tile Scintillator-Steel Hadronic Calorimeter. JINST, 18:P11018, 2022

    2022

  9. [9]

    JINST, 12:P10003, 2017

    CMS Collaboration.Particle-flow reconstruction and global event description with the CMS detector. JINST, 12:P10003, 2017

    2017

  10. [10]

    ATLAS Collaboration.Jet reconstruction and performance using particle flow with the ATLAS Detector. Eur. Phys. J. C, 77:466, 2017

    2017

  11. [11]

    Pinto.Shower Shapes in a Highly Granular SiPM-on-Tile Analog Hadron Calorimeter

    O. Pinto.Shower Shapes in a Highly Granular SiPM-on-Tile Analog Hadron Calorimeter. PhD thesis, Universit¨ at Hamburg, 2022.https://ediss.sub. uni-hamburg.de/handle/ediss/9855

    2022

  12. [12]

    Bieringer et al.Generative Models for Fast Simulation of Electromagnetic and Hadronic Showers in Highly Granular Calorimeters

    S. Bieringer et al.Generative Models for Fast Simulation of Electromagnetic and Hadronic Showers in Highly Granular Calorimeters. PoS, ICHEP2022:236, 2022

    2022

  13. [13]

    Diefenbacher et al.New angles on fast calorimeter shower simulation

    S. Diefenbacher et al.New angles on fast calorimeter shower simulation. Mach. Learn.: Sci. Technol., 4:035044, 2023

    2023

  14. [14]

    Buss et al.CaloHadronic: a diffusion model for the generation of hadronic showers

    T. Buss et al.CaloHadronic: a diffusion model for the generation of hadronic showers. JINST, 21:P01042, 2026

    2026

  15. [15]

    Wasserstein GAN

    M. Arjovskyet al.Wasserstein GAN. arXiv: 1701.07875v3 [stat.ML], 2017. 41

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  16. [16]

    Voloshynovskiy al.Information bottleneck through variational glasses

    S. Voloshynovskiy al.Information bottleneck through variational glasses. arXiv: 1912.00830v2 [cs.CV], 2019

    work page arXiv 1912

  17. [17]

    J. C. H. Rolph.Shower Separation in Five Dimensions Using Machine Learn- ing Techniques. PhD thesis, Universit¨ at Hamburg, 2023.https://ediss.sub. uni-hamburg.de/handle/ediss/10484

    2023

  18. [18]

    JINST, 16:T04002, 2021

    CMS HGCAL Collaboration.Construction and commissioning of CMS CE pro- totype silicon modules. JINST, 16:T04002, 2021

    2021

  19. [19]

    JINST, 16:T04001, 2021

    CMS HGCAL Collaboration.The DAQ system of the 12,000 channel CMS high granularity calorimeter prototype. JINST, 16:T04001, 2021

    2021

  20. [20]

    JINST, 11:P06013, 2016

    CALICE Collaboration.Hadron shower decomposition in the highly granular CALICE analogue hadron calorimeter. JINST, 11:P06013, 2016

    2016

  21. [21]

    Brianne.Time development of hadronic showers in a Highly Granular Analog Hadron Calorimeter

    E. Brianne.Time development of hadronic showers in a Highly Granular Analog Hadron Calorimeter. PhD thesis, Universit¨ at Hamburg, 2018.https://doi. org/10.3204/PUBDB-2018-02769

    work page doi:10.3204/pubdb-2018-02769 2018

  22. [22]

    Graf.Towards Precision Time and Energy Measurements in Highly Granular Hadronic Calorimeters

    C. Graf.Towards Precision Time and Energy Measurements in Highly Granular Hadronic Calorimeters. PhD thesis, Technische Universit¨ at M¨ unchen, 2020. https://mediatum.ub.tum.de/?id=1540800

    2020

  23. [23]

    Heuchel.Particle Flow Studies with Highly Granular Calorimeter Data

    D. Heuchel.Particle Flow Studies with Highly Granular Calorimeter Data. PhD thesis, Ruprecht-Karls-Universit¨ at Heidelberg, 2022.https://doi.org/ 10.11588/heidok.00031794. 42

    work page doi:10.11588/heidok.00031794 2022