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.

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

What is gammaVec here? #2

Open
coderaBruce opened this issue Jun 28, 2023 · 0 comments
Open

What is gammaVec here? #2

coderaBruce opened this issue Jun 28, 2023 · 0 comments

Comments

@coderaBruce
Copy link 

# note: gammaVec equals eta + diag(Lambda) in the Supplement
        gammaVec= L2const + XtXdiag * pdrop / (1.0-pdrop) # without zero-diag constraint            
        if zeroDiagConstraint:
            KK = VVt.T.dot(GG).dot(VVt)
            gammaVec= np.linalg.solve( CC * KK   , np.diag(KK))
        HH=CC *(-gammaVec) # Note: it is elementwise multiplication instead of matrix multiplication
        HH[ii_diag]+=1.0
        UU=HH.dot( VVt.T.dot(GG) )

Hi Herald, I am not sure if I should post my questions here, but any of your explanations would be extremely appreciated and helpful.
Based on the supplement of the paper, with no constraint, eta should be zero, and gemmaVec should be diag(XtX), but why do we add L2 constant here on gammaVec?
Additionally, why would we self increase the diagonal of HH by one? I do not think this appears on the formula...
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@coderaBruce