How to backpropagate the gradient of an error matrix for the weights in a custom layer using Tensorflow/Keras

Question

I am trying to build a custom layer that performs a noise injection to the weights and biases of a fully-connected layer. The code is shown below.

    def __init