A new diagnostic reveals that L=2 equivariant force field backbones preserve frequencies up to l=4 but collapse at l=5 on aspirin, consistent with a finite-degree span theorem and controls.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
other 1
citation-polarity summary
verdicts
UNVERDICTED 3representative citing papers
Regularized Newton's method for neural networks converges exponentially to zero loss with uniform spectral rates in the infinite-width limit via a derived Newton neural tangent kernel.
Neural ODE flow maps composed with embedding and projection yield shallow networks with universal approximation in C^0, preserved under separate Lipschitz or norm constraints but with quantified loss when both are imposed.
citing papers explorer
-
Diagnosing Spectral Ceilings in Equivariant Neural Force Fields
A new diagnostic reveals that L=2 equivariant force field backbones preserve frequencies up to l=4 but collapse at l=5 on aspirin, consistent with a finite-degree span theorem and controls.
-
Convergence Analysis of Newton's Method for Neural Networks in the Overparameterized Limit
Regularized Newton's method for neural networks converges exponentially to zero loss with uniform spectral rates in the infinite-width limit via a derived Newton neural tangent kernel.
-
Approximation properties of neural ODEs
Neural ODE flow maps composed with embedding and projection yield shallow networks with universal approximation in C^0, preserved under separate Lipschitz or norm constraints but with quantified loss when both are imposed.