End-to-end optimal detector design with mutual information surrogates

Article Article scientifique Article End-to-end optimal detector design with mutual information surrogates eng 2025-11-17 2025 author Wozniak Kinga Geneva U. 410864 Département de physique nucléaire et corpusculaire (FINUC) author Mulligan Stephen Geneva U. 417365 Département de physique nucléaire et corpusculaire (FINUC) author Kieseler Jan KIT, Karlsruhe author Klute Markus KIT, Karlsruhe author Fleuret François Jean U. Geneva (main) 766513 Département d'informatique (DINFO) author Golling Tobias Geneva U. 827102 Département de physique nucléaire et corpusculaire (FINUC) 045047 10.1088/2632-2153/ae1acb 2632-2153 2503.14342 https://iopscience.iop.org/article/10.1088/2632-2153/ae1acb We introduce a novel approach for end-to-end black-box optimization of high energy physics (HEP) detectors using local deep learning surrogates. These surrogates approximate a scalar objective function that encapsulates the complex interplay of particle-matter interactions and physics analysis goals. In addition to a standard reconstruction-based metric commonly used in the field, we investigate the information-theoretic metric of mutual information. In contrast to traditional methods, mutual information is inherently task-agnostic, offering a broader optimization paradigm that is less constrained by predefined targets. We demonstrate the effectiveness of our method in a realistic physics analysis scenario: optimizing the thicknesses of calorimeter detector layers based on simulated particle interactions. The surrogate model learns to approximate objective gradients, enabling efficient optimization with respect to energy resolution. Our findings reveal three key insights: (1) end-to-end black-box optimization using local surrogates is a practical and compelling approach for detector design, providing direct optimization of detector parameters in alignment with physics analysis goals; (2) mutual information-based optimization yields design choices that closely match those from state-of-the-art physics-informed methods, indicating that these approaches operate near optimality and reinforcing their reliability in HEP detector design; and (3) information-theoretic methods provide a powerful, generalizable framework for optimizing scientific instruments. By reframing the optimization process through an information-theoretic lens rather than domain-specific heuristics, mutual information enables the exploration of new avenues for discovery beyond conventional approaches. Detector design Deep learning Surrogate Mutual information Physics High-energy Dewey 500.2 Alexander von Humboldt-Stiftung Swiss National Science Foundation At the two upgrade frontiers: machine learning and the ITk Pixel detector 212127 CH 34 Département de physique nucléaire et corpusculaire 132 false Machine learning: science and technology 6 4 021273b6-3bf7-4a58-915c-800a1d32c91c Article (Published version) Public file.pdf 4289319 application/pdf CC BY-4.0 bce46eadb136bd0923d67394565c144dac600f58cdb28b6a2b4699be4a2bee49 openAccess Faculté des sciences / Section de physique / Département de physique nucléaire et corpusculaire

Title	Published in	Access level	OA Policy	Year	Views	Downloads

End-to-end optimal detector design with mutual information surrogates	Machine learning: science and technology			2025	5	6

Archive ouverte UNIGE

Wozniak, Kinga