A neural network augmented with the geometric mean of multiple small parameters approximates solutions to singularly perturbed dynamical systems with satisfactory accuracy on tested coupled cases.
On the eigenvector bias of Fourier feature networks: from regression to solving multi-scale PDEs with physics-informed neural networks.Comput
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.NA 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Two-scale Neural Networks for Singularly Perturbed Dynamical Systems with Multiple Parameters
A neural network augmented with the geometric mean of multiple small parameters approximates solutions to singularly perturbed dynamical systems with satisfactory accuracy on tested coupled cases.