Efficient Neural Surrogates for Linear Radiation Transport on the Lattice and Hohlraum benchmarks
Abstract
Linear radiation transport equations (RTEs) form the simulation foundations underpinning design and analysis tasks in nuclear engineering, inertial confinement fusion, medical imaging, and astrophysics, but resolving the high-dimensional phase space at engineering fidelity remains expensive enough that outer-loop workflows, such as design optimization, uncertainty quantification, and parameter sweeps, are routinely budget-bound on traditional solvers. Neural surrogates promise to relax this bottleneck by amortizing simulation cost across thousands of downstream queries, but the architectural choices and engineered inductive biases that make a surrogate accurate on one transport problem do not transfer straightforwardly across model families. We benchmark two parameter-matched neural surrogate architectures, the physics-attention Transolver and the multi-scale graph network Bi-Stride Multi-Scale MeshGraphNet (BSMS-MGN), as end-to-end approximations of the final-time particle concentration for the two-dimensional linear RTE on the canonical Lattice and Hohlraum benchmarks. An ablation across Fourier features and region-weighted training loss exposes strongly architecture-dependent inductive-bias preferences, indicating that design choices common to physics-informed surrogate workflows must be revisited per architecture rather than imported across model families, and that downstream utility depends on per-QoI sensitivity rather than a single field-level score. The model training recipe, training data, and evaluation pipeline are released alongside this paper to support reproduction, transfer to related transport problems, and evaluation as amortized forward-model components in larger outer-loop simulation workflows.
Related
- Aluminum wide-range EOS sensitivity in Z-pinch dynamic-hohlraum liner-foam impact and outgoing radiation
- Anomalous enhancement of thermal radiation transport by quasidisorder
- Efficient Transport Networks through Edge Pruning by Relative Resistance
- Efficient simulations of spectra and transport for multiorbital quantum impurities
- Efficient, Geometry-only Prediction of Advection-dominated Particle Transport through Saturated Soils
BibTeX
@misc{gonzales2026efficient,
title = {{Efficient Neural Surrogates for Linear Radiation Transport on the Lattice and Hohlraum benchmarks}},
author = {Carmelo Gonzales and Steffen Schotthöfer and Cory D. Hauck},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.04665},
eprint = {2610.04665},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.04665},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Efficient Neural Surrogates for Linear Radiation Transport on the Lattice and Hohlraum benchmarks
AU - Carmelo Gonzales
AU - Steffen Schotthöfer
AU - Cory D. Hauck
PY - 2026
DA - 2026/10/03
PB - arXiv
DO - 10.48550/arxiv.2610.04665
UR - https://arxiv.org/abs/2610.04665
AB - Linear radiation transport equations (RTEs) form the simulation foundations underpinning design and analysis tasks in nuclear engineering, inertial confinement fusion, medical imaging, and astrophysics, but resolving the high-dimensional phase space at engineering fidelity remains expensive enough that outer-loop workflows, such as design optimization, uncertainty quantification, and parameter sweeps, are routinely budget-bound on traditional solvers. Neural surrogates promise to relax this bottleneck by amortizing simulation cost across thousands of downstream queries, but the architectural choices and engineered inductive biases that make a surrogate accurate on one transport problem do not transfer straightforwardly across model families. We benchmark two parameter-matched neural surrogate architectures, the physics-attention Transolver and the multi-scale graph network Bi-Stride Multi-Scale MeshGraphNet (BSMS-MGN), as end-to-end approximations of the final-time particle concentration for the two-dimensional linear RTE on the canonical Lattice and Hohlraum benchmarks. An ablation across Fourier features and region-weighted training loss exposes strongly architecture-dependent inductive-bias preferences, indicating that design choices common to physics-informed surrogate workflows must be revisited per architecture rather than imported across model families, and that downstream utility depends on per-QoI sensitivity rather than a single field-level score. The model training recipe, training data, and evaluation pipeline are released alongside this paper to support reproduction, transfer to related transport problems, and evaluation as amortized forward-model components in larger outer-loop simulation workflows.
ER -