Physics informed neural network #3087
Nikolaos-Matthaiakakis
started this conversation in
Ideas
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hello,
It might be an interesting idea to test a new feature that would combine Physics-Informed Neural Networks (PINNs) with Tidy3D’s FDTD simulations. The PINNs can learn from FDTD data (since your severs run simulations very fast) and also be trained to respect Maxwell’s equations (or even be used for multi-physics problems). This could complement the adjoint method as an alternative with diffferent advantages and disadvantages.
Best wishes
Nikolaos
Beta Was this translation helpful? Give feedback.
All reactions