Regarding GAN and its loss objective

How to use 'mse' as loss function?

In the example provided, sigmoid is used. How to modify the function to make the loss objective as minimum mse.

Answer 1

[Sourav Bairagya]

You can use 'mse' function from Deep Learning Toolbox which calculates half mean sqaured error between given two inputs.

dlY = mse(dlX,targets);

Format for 'dlX' or 'targets' will be as follows: [height, width, no of channels, no of observations].

'dlX' and 'targets' can be of datatype dlarray or numeric array.

If you want to use whole mse loss, then multiply the output with 2, i.e.

dlY = 2*mse(dlX,targets);

For more information you can leverage this link:

https://www.mathworks.com/help/deeplearning/ref/dlarray.mse.html?searchHighlight=mse&s_tid=doc_srchtitle