English

Divergence-free Linearized Neural Networks: Integral Representation and Optimal Approximation Rates

Numerical Analysis 2026-03-31 v1 Numerical Analysis

Abstract

This paper studies the numerical approximation of divergence-free vector fields by linearized shallow neural networks, also referred to as random feature models or finite neuron spaces. Combining the stable potential lifting for divergence-free fields with the scalar Sobolev integral representation theory via ReLUk^k networks, we derive a core integral representation of divergence-free Sobolev vector fields through antisymmetric potentials parameterized by linearized ReLUk^k neural networks. This representation, together with a quasi-uniform distribution argument for the inner parameters, yields optimal approximation rates for such linearized ReLUk^k neural networks under an exact divergence-free constraint. Numerical experiments in two and three spatial dimensions, including L2L^2 projection and steady Stokes problems, confirm the theoretical rates and illustrate the effectiveness of exactly divergence-free conditions in computation.

Keywords

Cite

@article{arxiv.2603.28638,
  title  = {Divergence-free Linearized Neural Networks: Integral Representation and Optimal Approximation Rates},
  author = {Juncai He and Xinliang Liu and Zitong Tian},
  journal= {arXiv preprint arXiv:2603.28638},
  year   = {2026}
}

Comments

27 pages, 11 figures

R2 v1 2026-07-01T11:44:24.741Z