A unified deep artificial neural network approach to partial differential equations in complex geometries
A unified deep artificial neural network approach to partial differential equations in complex geometries
Jens Berg,K. Nyström
2017 · DOI: 10.1016/j.neucom.2018.06.056
Neurocomputing · 631 Citations
TLDR
This paper shows how to modify the backpropagation algorithm to compute the partial derivatives of the network output with respect to the space variables which is needed to approximate the differential operator.