"Recurrent Neural Networks as Optimal Mesh Refinement Strategies"Feischl, MichaelWe show that an optimal finite element mesh refinement algorithm for a prototypical ellipticPDE can be learned by a recurrent neural network with a fixed number of trainable parameters independent of the desired accuracy and the input size, i.e., number of elements of the mesh. |
https://arxiv.org/pdf/1909.04275.pdf |
« back