Advection Diffusion Recurrent Convolutional Network

Implementation of "Observations on Recurrent Loss in the Neural Network Model of a [PDE]: the Advection-Diffusion Equation"

This is an implementation of the algorithms discussed in "Observations on Recurrent Loss in the Neural Network Model of a Partial Differential Equation: the Advection-Diffusion Equation". The function "Advection_Diffusion_Recursive_Convolutional_Network_Wrapper" generates all or a portion of the results for test cases outlined in the paper. The function "Advection_Diffusion_Recursive_Convolutional_Network_Test" constructs a structure "Network" containing network and problem parameters. The network is trained and evaluated using the functions "Apply_Network", "Backpropagation" and "Evaluate_Network".