Toward the end-to-end optimization of particle physics instruments with differentiable programming

Dorigo, T.; Giammanco, A.; Vischia, P.; Max, Aehle,; Bawaj, M.; Boldyrev, A. S.; Manzano, P. De Castro; Деркач, Д.; Donini, J.; Edelen, Auralee; Fanzago, F.; Gauger, Nicolas R.; Gläser, Christian; Baydin, Atılım G.; Heinrich, L.; Kieseler, J.; Krause, Claudius; Lagrange, Maxime; Lamparth, Max; Layer, L.; Maier, G.; Nardi, Federico; Pettersen, Helge E.S.; Ramos, Alberto; Ratnikov, F.; Röhrich, D.; Austri, Roberto Ruiz de; Arbol, P. Martínez Ruiz del; Savchenko, Oleg; Simpson, Nathan Daniel; Strong, G.; Taliercio, A.; Tosi, M.; Ustyuzhanin, A.; Zaraket, H.

doi:10.1016/j.revip.2023.100085

Cited by 16 publications

(14 citation statements)

References 187 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This research is part of a larger movement to automate detector design with machine learning. Recent works have considered a number of different instruments with a variety of approaches [29][30][31][32][33][34]. The studies presented here could be combined with a point cloud generative model for an end-to-end optimization [35][36][37][38][39][40][41].…”

Section: Discussionmentioning

confidence: 99%

The optimal use of segmentation for sampling calorimeters

Torales Acosta,

Karki,

Karande

et al. 2024

J. Inst.

View full text Add to dashboard Cite

One of the key design choices of any sampling calorimeter is how fine to make the longitudinal and transverse segmentation. To inform this choice, we study the impact of calorimeter segmentation on energy reconstruction. To ensure that the trends are due entirely to hardware and not to a sub-optimal use of segmentation, we deploy deep neural networks to perform the reconstruction. These networks make use of all available information by representing the calorimeter as a point cloud. To demonstrate our approach, we simulate a detector similar to the forward calorimeter system intended for use in the ePIC detector, which will operate at the upcoming Electron Ion Collider. We find that for the energy estimation of isolated charged pion showers, relatively fine longitudinal segmentation is key to achieving an energy resolution that is better than 10% across the full phase space. These results provide a valuable benchmark for ongoing EIC detector optimizations and may also inform future studies involving high-granularity calorimeters in other experiments at various facilities.

show abstract

Section: Discussionmentioning

confidence: 99%

The optimal use of segmentation for sampling calorimeters

Torales Acosta,

Karki,

Karande

et al. 2024

J. Inst.

View full text Add to dashboard Cite

show abstract

“…Finally, image-or point-cloud-based generative models often use EMD to evaluate the fidelity of generated data [40][41][42][43], and using a differentiable approximation would enable direct minimization during training [44]. By directly optimizing EMD, better and more computationally efficient physics simulators can be trained, potentially enabling end-to-end detector optimization via differentiable programming [45].…”

Section: Discussionmentioning

confidence: 99%

Differentiable Earth mover’s distance for data compression at the high-luminosity LHC

Shenoy,

Duarte,

Herwig

et al. 2023

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

The Earth mover’s distance (EMD) is a useful metric for image recognition and classification, but its usual implementations are not differentiable or too slow to be used as a loss function for training other algorithms via gradient descent. In this paper, we train a convolutional neural network (CNN) to learn a differentiable, fast approximation of the EMD and demonstrate that it can be used as a substitute for computing-intensive EMD implementations. We apply this differentiable approximation in the training of an autoencoder-inspired neural network (encoder NN) for data compression at the high-luminosity LHC at CERN The goal of this encoder NN is to compress the data while preserving the information related to the distribution of energy deposits in particle detectors. We demonstrate that the performance of our encoder NN trained using the differentiable EMD CNN surpasses that of training with loss functions based on mean squared error.

show abstract

“…In contrast, our work heads towards the long-time goal of finding optimal parameters of a pCT setup, including geometric parameters, material compositions, and algorithmic parameters of the reconstruction software. In this context, a meaningful objective function is given by the error of the reconstructed RSP image compared with the original RSP used in the Monte-Carlo simulation (Dorigo et al 2023), averaged over a collection of such original RSP images.…”

Section: Gradient-based Optimizationmentioning

confidence: 99%

“…Surrogate models for calorimeter showers are a very active field of research, see e.g. Dorigo et al (2023).…”

Section: Differentiation Of Randomized Codementioning

confidence: 99%

“…Algorithmic differentiation (AD) Walther 2008, Naumann 2011) is a set of techniques to efficiently obtain precise derivatives of a mathematical function given by a computer program. Such derivatives have been successfully used to solve optimization problems in various contexts, such as machine learning (ML) (Baydin et al 2017) and computational fluid dynamics (CFD) (Albring et al 2016), and AD is currently also adopted in the fundamental physics community for detector optimization (Baydin et al 2021, Strong et al 2022a, 2022b, Dorigo et al 2023. In all of these applications, an objective or loss function is formulated, which maps…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Exploration of differentiability in a proton computed tomography simulation framework

Aehle,

Alme,

Gábor Barnaföldi

et al. 2023

Phys. Med. Biol.

Self Cite

View full text Add to dashboard Cite

Objective. Gradient-based optimization using algorithmic derivatives can be a useful technique to improve engineering designs with respect to a computer-implemented objective function. Likewise, uncertainty quantification through computer simulations can be carried out by means of derivatives of the computer simulation. However, the effectiveness of these techniques depends on how “well-linearizable” the software is. In this study, we assess how promising derivative information of a typical proton computed tomography (pCT) scan computer simulation is for the aforementioned applications. Approach. This study is mainly based on numerical experiments, in which we repeatedly evaluate three representative computational steps with perturbed input values. We support our observations with a review of the algorithmic steps and arithmetic operations performed by the software, using debugging techniques. Main results.The model-based iterative reconstruction (MBIR) subprocedure (at the end of the software pipeline) and the Monte Carlo (MC) simulation (at the beginning) were piecewise differentiable. However, the observed high density and magnitude of jumps was likely to preclude most meaningful uses of the derivatives. Jumps in the MBIR function arose from the discrete computation of the set of voxels intersected by a proton path, and could be reduced in magnitude by a “fuzzy voxels” approach. The investigated jumps in the MC function arose from local changes in the control flow that affected the amount of consumed random numbers. The tracking algorithm solves an inherently non-differentiable problem. Significance. Besides the technical challenges of merely applying AD to existing software projects, the MC and MBIR codes must be adapted to compute smoother functions. For the MBIR code, we presented one possible approach for this while for the MC code, this will be subject to further research. For the tracking subprocedure, further research on surrogate models is necessary.

show abstract

Toward the end-to-end optimization of particle physics instruments with differentiable programming

Cited by 16 publications

References 187 publications

The optimal use of segmentation for sampling calorimeters

The optimal use of segmentation for sampling calorimeters

Differentiable Earth mover’s distance for data compression at the high-luminosity LHC

Exploration of differentiability in a proton computed tomography simulation framework

Contact Info

Product

Resources

About