sign on loss function

[johnwlambert]

Hi, in the paper pseudocode, for Loss_d and Loss_g, the gradient ascent is turned into gradient descent by making the whole loss negative (negative signs in front of every term).

However, in the code, I don't see any of those negative signs. What am I missing?

Thank you!

[IshmaelBelghazi]

The negative term in front of the discriminator loss is factored in the softplus.

[edgarriba]

@johnwlambert check the expanded formulas

[edgarriba]

whereas softplus(x) = log(1 + exp(x))