-
Notifications
You must be signed in to change notification settings - Fork 298
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
A question about adversarial loss #38
Comments
This point has also triggered my attention. The change has occured between SEANet that introduced For my case with EBEN, I just figured out that this criterion was working well (better than LSGAN, WGAN, classic GAN, or original geometric GAN formulation). A possible explanation may be the symmetric use case of the discriminator that should only output values in the range [-1,1], helping to stabilize training by avoiding overconfidence. |
@jhauret it's worth noting that BigVGAN, which is also SOTA, uses an LSGAN loss. I am not aware that the discriminators only output values in the range [-1, 1]. Why do you say that? It appears to me that many discriminators do not apply a squashing function at the last layer, in order to avoid vanishing gradient to the generator. With that said, to answer @BakerBunker's question why |
Thanks for pointing out the use of LSGAN loss in BigVGAN. Sorry if I was unclear. In fact, the values of the discriminators can be outside [-1, 1], but if you minimize You are also right about your last point, but this loss change has been seen in other papers before such a loss balancer was introduced. |
❓ Questions
In the paper 3.4 "Discriminative Loss" section, adversarial loss is constructed as$l_g(\hat{x})=\mathbb{E}[max(0,1-D_k(\hat{x}))]$ , but in the original hinge loss paper, adversarial loss is constructed as $-\mathbb{E}[D(\hat{x})]$ .
So I want to know, why the adversarial loss in this paper is different from the original hinge loss?
The text was updated successfully, but these errors were encountered: