End-to-End Optimal Detector Design with Mutual Information Surrogates
Kinga Anna Woźniak, Stephen Mulligan, Jan Kieseler, Markus Klute, François Fleuret, Tobias Golling · 2025
Read End-to-End Optimal Detector Design with Mutual Information Surrogates by Kinga Anna Woźniak, Stephen Mulligan, Jan Kieseler, Markus Klute, François Fleuret, Tobias Golling (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-14342.txt.
Description
Kinga Anna Woźniak and co-authors present a way to design particle detectors end to end in a 2025 arXiv paper. Detector design must balance matter interactions against analysis goals, a tangled trade-off. Their method optimizes the whole chain with local deep learning surrogates.
The surrogates learn a scalar objective and its gradients, enabling fast black-box tuning. Beside the standard reconstruction metric, they test mutual information as a task-free objective. The demo tunes calorimeter layer thicknesses from simulated particle hits. Mutual information designs closely matched physics-informed ones. The authors read this as a sign both sit near the optimum.
Catalog record
| CREATOR | Kinga Anna Woźniak, Stephen Mulligan, Jan Kieseler, Markus Klute, François Fleuret, Tobias Golling |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-14342