A continuous data assimilation framework enables a-priori training of solver-conditioned neural turbulence closures that remain stable at deployment and track discretization errors.
What you see is not what you get: Neural partial differential equations and the illusion of learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.flu-dyn 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Deep Learning of Solver-Aware Turbulence Closures from Nudged LES Dynamics
A continuous data assimilation framework enables a-priori training of solver-conditioned neural turbulence closures that remain stable at deployment and track discretization errors.