-
Notifications
You must be signed in to change notification settings - Fork 101
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
Implementation of Asymmetric Clipping #10
Comments
Notice the definition in the code:
so basically: this is done to prevent calculating again and again (1-p) along the code... |
Oh I see. Nice trick! Thanks for the quick response. I'll close the issue now. |
Closed
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Thanks for such an interesting paper 👍
In the paper's equation (4), asymmetric probability shifting is
p_m = max(p-m, 0)
, but in the implementation, it's called asymmetric clipping and there isxs_neg = (xs_neg + self.clip).clamp(max=1)
which is probablyp_m = min(p+m, 1)
.Is there a reason for this difference?
The text was updated successfully, but these errors were encountered: