Discriminator dominates over generator
See original GitHub issueI could get audible results at epoch>350, but they don’t look good. Also, d_loss gets too low and g_loss gets too high.

Perhaps this could be caused by:
- Discriminator may have learned that real data have discrete values (
np.int16
): Adding a gaussian noise may help. See https://github.com/soumith/ganhacks/issues/14
Issue Analytics
- State:
- Created 4 years ago
- Comments:25 (14 by maintainers)
Top Results From Across the Web
Common Problems | Machine Learning - Google Developers
Each iteration of generator over-optimizes for a particular discriminator, and the discriminator never manages to learn its way out of the trap.
Read more >How to Identify and Diagnose GAN Failure Modes
Discriminator accuracy on real and fake images is expected to sit around 80%. Variance of generator and discriminator loss is expected to remain ......
Read more >GANs Failure Modes: How to Identify and Monitor Them
An aggressive modifier is bad news for training GANs. It results in the inability to find the equilibrium between generator loss and discriminator...
Read more >Why use the output of the generator to train the discriminator ...
The main reason that the discriminator is trained concurrently with the generator is to provide (at least in theory) a smooth and gradual ......
Read more >python - How to balance the generator and the discriminator ...
that is, the gradients on the Generator are higher and start to decrease after a while, and in the meanwhile the gradients on...
Read more >
Top Related Medium Post
No results found
Top Related StackOverflow Question
No results found
Troubleshoot Live Code
Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free
Top Related Reddit Thread
No results found
Top Related Hackernoon Post
No results found
Top Related Tweet
No results found
Top Related Dev.to Post
No results found
Top Related Hashnode Post
No results found
This is getting interesting! After about 2000 epochs of training, discriminator finds it almost impossible to discriminate real/fake.
Getting good quality audio while generator loss is still going up, too. (Audio samples at http://swpark.me/melgan/)
@rishikksh20 Even if generator loss is going up, it doesn’t seem that discriminator is overly better than generator. How do you think? Shall we close this issue?