A block-diagonal symmetrizer and algebraic conditions on closure blocks enable a data-learnable parametrization of ML moment closures for 2D RTE that guarantees symmetrizable hyperbolicity by construction.
Physics informed neural networks for simulating radiative transfer.Journal of Quantitative Spectroscopy and Radiative Transfer, 270:107705, 2021
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
math.NA 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
MgNet learns the RTE solution operator via a multigrid-inspired architecture with neural sub-operators and adaptive angular filtering, delivering at least 10x speedup as a preconditioner with generalization to new parameters.
citing papers explorer
-
Machine learning moment closure models for the radiative transfer equation IV: enforcing symmetrizable hyperbolicity in two dimensions
A block-diagonal symmetrizer and algebraic conditions on closure blocks enable a data-learnable parametrization of ML moment closures for 2D RTE that guarantees symmetrizable hyperbolicity by construction.
-
A Filtered MgNet Solver For Radiative Transfer Equations
MgNet learns the RTE solution operator via a multigrid-inspired architecture with neural sub-operators and adaptive angular filtering, delivering at least 10x speedup as a preconditioner with generalization to new parameters.