Skip to content
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

Question on architecture #22

Open
ovcGIT opened this issue Dec 24, 2024 · 0 comments
Open

Question on architecture #22

ovcGIT opened this issue Dec 24, 2024 · 0 comments

Comments

@ovcGIT
Copy link

ovcGIT commented Dec 24, 2024

Few clarifications/questions (i'm a beginner, please pardon any basic questions or redirect if an alternate channel is more appropriate)

  1. Visualization diagram on GIT: Input gate diagram visualization shows a Ct, is that actually Ct tilde from the paper?
  2. Visualization diagram on GIT: There is a tanh activation against Ct in the diagram visualization, is that actually a sigmoid from the paper (text below Eq 15) ?
  3. Fig 2 in paper: can you help clarify what are the inputs (red circles on extreme left of Fig 2) - because only one of them has a sigmoid activation. From the paper it seemed like both input / forget had activations if these are what they denote)
  4. Fig 2in paper: The KAN layers (called sub-layers) in the paper are these non-recurrent across time steps and NOT connected to each other
  5. Fig 3 in paper: Three TKAN layers: looks these not connected? are the outputs simply concatenated
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant