Progress in End-to-End Optimization of Detectors for Fundamental Physics with Differentiable Programming
Abstract
In this article we examine recent developments in the research area concerning the creation of end-to-end models for the complete optimization of measuring instruments. The models we consider rely on differentiable programming methods and on the specification of a software pipeline including all factors impacting performance -- from the data-generating processes to their reconstruction and the extraction of inference on the parameters of interest of a measuring instrument -- along with the careful specification of a utility function well aligned with the end goals of the experiment. Building on previous studies originated within the MODE Collaboration, we focus specifically on applications involving instruments for particle physics experimentation, as well as industrial and medical applications that share the detection of radiation as their data-generating mechanism.
Cite
@article{arxiv.2310.05673,
title = {Progress in End-to-End Optimization of Detectors for Fundamental Physics with Differentiable Programming},
author = {Max Aehle and Lorenzo Arsini and R. Belén Barreiro and Anastasios Belias and Florian Bury and Susana Cebrian and Alexander Demin and Jennet Dickinson and Julien Donini and Tommaso Dorigo and Michele Doro and Nicolas R. Gauger and Andrea Giammanco and Lindsey Gray and Borja S. González and Verena Kain and Jan Kieseler and Lisa Kusch and Marcus Liwicki and Gernot Maier and Federico Nardi and Fedor Ratnikov and Ryan Roussel and Roberto Ruiz de Austri and Fredrik Sandin and Michael Schenk and Bruno Scarpa and Pedro Silva and Giles C. Strong and Pietro Vischia},
journal= {arXiv preprint arXiv:2310.05673},
year = {2025}
}
Comments
70 pages, 17 figures. To be submitted to journal