NTK neural networks achieve minimax optimal adversarial regression rates in Sobolev spaces using gradient flow with early stopping, but minimum norm interpolants are vulnerable in the overfitting regime.
Moritz Haas, David Holzmüller, Ulrike von Luxburg, and Ingo Steinwart
3 Pith papers cite this work, alongside 959 external citations. Polarity classification is still indexing.
3
Pith papers citing it
959
external citations · OpenAlex
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Establishes boundedness for the regularized bilinear cone multiplier on R^2 x R^2 via square function decomposition, maximal function estimates, and geometric methods from Córdoba and Carbery.
Spectral asymptotics for negative fractional powers of hypoelliptic operators on graded Lie groups generalize Birman-Solomyak and imply a version of Connes' integration formula.
citing papers explorer
-
Adversarial Robustness of NTK Neural Networks
NTK neural networks achieve minimax optimal adversarial regression rates in Sobolev spaces using gradient flow with early stopping, but minimum norm interpolants are vulnerable in the overfitting regime.
-
The bilinear cone multiplier on $\mathbb{R}^2\times \mathbb{R}^2$
Establishes boundedness for the regularized bilinear cone multiplier on R^2 x R^2 via square function decomposition, maximal function estimates, and geometric methods from Córdoba and Carbery.
-
Weyl asymptotic formulas in the nilpotent Lie group setting
Spectral asymptotics for negative fractional powers of hypoelliptic operators on graded Lie groups generalize Birman-Solomyak and imply a version of Connes' integration formula.