The paper claims spline-parameterized KAN least-squares estimators achieve the minimax univariate regression rate O(n^{-2r/(2r+1)}) for additive and multiplicative KAN structures, independent of dimension.
On deep learning as a remedy for the curse of dimensionality in nonparametric regression
1 Pith paper cite this work, alongside 247 external citations. Polarity classification is still indexing.
1
Pith paper citing it
247
external citations · OpenAlex
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
On the Rate of Convergence of Kolmogorov-Arnold Network Regression Estimators
The paper claims spline-parameterized KAN least-squares estimators achieve the minimax univariate regression rate O(n^{-2r/(2r+1)}) for additive and multiplicative KAN structures, independent of dimension.