Exploring the Boundaries of Differentiable Radiation Transport and Detector Simulation

MCML Authors

Lukas Heinrich

Prof. Dr.

Collaborating PI

Data Science in Physics

Abstract

We present an application of automatic differentiation for particle transport through matter using a Geant4-like radiation transport simulation with a full electromagnetic physics model. When differentiating this step-based transport, we observe exploding gradients driven by rare but extreme sensitivities at material boundaries, which propagate through subsequent transport and shower development. To obtain usable derivatives for optimization, we introduce a targeted mitigation strategy that stops gradient propagation through boundary-crossing operations under identifiable unstable conditions while leaving the forward (primal) simulation unchanged. We demonstrate that this enables stable, optimization-ready gradients in a detector-design problem.

misc KZN+26

---

Preprint

May. 2026

---

Authors

J. Krupa • Y. Zhao • M. Novak • M. Aehle • M. Sagebaum • L. Chen • N. Gauger • M. Liu • L. Heinrich • M. Kagan

---

Links

arXiv GitHub

---

Research Area

 C3 | Physics and Geo Sciences