Regularized implicit Runge-Kutta integrators for nonlinear parametrizations such as neural networks are shown to have global errors bounded by the non-parametric integrator's error plus the size of the residuals.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.NA 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Regularized dynamical parametric approximation of stiff evolution problems
Regularized implicit Runge-Kutta integrators for nonlinear parametrizations such as neural networks are shown to have global errors bounded by the non-parametric integrator's error plus the size of the residuals.