Skip to content

Fix ACGAN training #23

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.

Sign up for GitHub

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

Jump to bottom
Open
yang-song wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Fix ACGAN training #23

yang-song wants to merge 1 commit into from

Conversation

@yang-song
Copy link

@yang-song yang-song commented Mar 18, 2018

The ACGAN training procedure is not optimal
@yang-song
Fix ACGAN training
2811ad1 
The ACGAN training procedure is not optimal
@ElApseR
Copy link

ElApseR commented Feb 6, 2019

I think this is correct and makes training stable.

By the article,

$$L_s=E[logP(S=real|X_{real})]+E[logP(S=fake|X_{fake})]$$ $$L_c=E[logP(C=c|X_{real})]+E[logP(C=c|X_{fake})]$$ $$D = argmax(L_s+L_c)$$ $$G = argmax(L_c-L_s)$$

which means, you don't need to train Q separately. Also, Q shouldn't optimize variables of G.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

2 participants

@yang-song @ElApseR