KIV regression converges in strong L2 norm and attains minimax optimal rates over smoothness classes, with improved rates from general spectral regularization in stage one.
If we fix the total distance in the Landweber iteration to ξ−1 ∶= τ k and take τ → 0+ , we obtain the gradient flow filter function gξ( x) = ( 1 −e−x ξ ) x−1
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Nonparametric Instrumental Regression via Kernel Methods is Minimax Optimal
KIV regression converges in strong L2 norm and attains minimax optimal rates over smoothness classes, with improved rates from general spectral regularization in stage one.