-
-
Notifications
You must be signed in to change notification settings - Fork 70
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
Reverse-mode AD for SDEs #1084
Labels
Comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Reverse adjoints for SDEs only works with 'TrackerAdjoint()' and only on CPU. 🐞
Training Large (e.g, Neural) SDEs on GPUs fails. The only working solution is 'TrackerAdjoint()' and this only currently works on CPU.
None of the continuous adjoints methods, e.g. 'InterpolatingAdjoint()' or 'BackwardsolveAdjoint()' work either on cpu or gpu.
MWE
Error & Stacktrace
I am using the latest releases for the packages and Julia 1.10.4.
The text was updated successfully, but these errors were encountered: