The role of parameter Jacobians in the stability of network outputs
Abstract
In the framework of network dynamics, learning models, and neural tangent kernels (NTK), we show that the corresponding linearized dynamics leads naturally to a semigroup formulation. More precisely, in our analysis of input/output models, the time-dynamics is presented via special semigroups of linear operators on Hilbert spaces, together with an associated class of semigroup perturbations. In this context, we then present new and explicit a priori perturbation-bound results: for the fixed-kernel linearization constructions arising in the NTK setting, we prove norm-bounds on the corresponding semigroup perturbations, in the form of explicit finite-time perturbation estimates. We further present refinements on prescribed task spaces, Cesàro-averaged (ergodic) comparisons estimates, and versions in which the lower spectral edge assumption is replaced by a spectral-distribution condition. We also extend the comparison to nonautonomous NTK evolutions through piecewise-frozen approximations, record a corresponding discrete Euler specialization, and offer worked examples in order to illustrate our perturbation-bound estimates.
Related
- Operator-Theoretic Stability and Observer Synthesis for Parameter-Dependent Vlasov--Maxwell Dynamics
- Experimental and theoretical investigation of parameter-driven stability and chaos in a non-linear neuron circuit model
- Multi-Objective Fractional-Order Control Design for Solar Collector Systems Under Parameter Perturbations and Stability Constraints
- Harmonizable Multifractional Stable Motion: estimators for the Hurst function and the stability parameter from discrete-time observations, their rates of convergence and the associated central limit theorems
- Optimal Tradeoffs Between Network Size and Parameter Magnitude in Neural Approximation and Minimax Regression
BibTeX
@misc{jeong2026role,
title = {{The role of parameter Jacobians in the stability of network outputs}},
author = {Halyun Jeong and Palle E. T. Jorgensen and Hyun-Kyoung Kwon and Myung-Sin Song and James Tian},
howpublished = {arXiv},
year = {2026},
month = aug,
doi = {10.48550/arxiv.2608.27748},
eprint = {2608.27748},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.27748},
}
RIS
TY - UNPB
TI - The role of parameter Jacobians in the stability of network outputs
AU - Halyun Jeong
AU - Palle E. T. Jorgensen
AU - Hyun-Kyoung Kwon
AU - Myung-Sin Song
AU - James Tian
PY - 2026
DA - 2026/08/27
PB - arXiv
DO - 10.48550/arxiv.2608.27748
UR - https://arxiv.org/abs/2608.27748
AB - In the framework of network dynamics, learning models, and neural tangent kernels (NTK), we show that the corresponding linearized dynamics leads naturally to a semigroup formulation. More precisely, in our analysis of input/output models, the time-dynamics is presented via special semigroups of linear operators on Hilbert spaces, together with an associated class of semigroup perturbations. In this context, we then present new and explicit a priori perturbation-bound results: for the fixed-kernel linearization constructions arising in the NTK setting, we prove norm-bounds on the corresponding semigroup perturbations, in the form of explicit finite-time perturbation estimates. We further present refinements on prescribed task spaces, Cesàro-averaged (ergodic) comparisons estimates, and versions in which the lower spectral edge assumption is replaced by a spectral-distribution condition. We also extend the comparison to nonautonomous NTK evolutions through piecewise-frozen approximations, record a corresponding discrete Euler specialization, and offer worked examples in order to illustrate our perturbation-bound estimates.
ER -