Robustness of Deep ReLU Networks to Misclassification of High-Dimensional Data

Věra Kůrková

Robustness of Deep ReLU Networks to Misclassification of High-Dimensional Data

Machine Learning 2026-02-24 v1 Neural and Evolutionary Computing

Authors: Věra Kůrková

Abstract

We present a theoretical study of the robustness of parameterized networks to random input perturbations. Specifically, we analyze local robustness at a given network input by quantifying the probability that a small additive random perturbation of the input leads to misclassification. For deep networks with rectified linear units, we derive lower bounds on local robustness in terms of the input dimensionality and the total number of network units.

Keywords

deep neural networks ReLU robustness misclassification high-dimensional data

Cite

@article{arxiv.2602.18674,
  title  = {Robustness of Deep ReLU Networks to Misclassification of High-Dimensional Data},
  author = {Věra Kůrková},
  journal= {arXiv preprint arXiv:2602.18674},
  year   = {2026}
}

Comments

15 pages, 4 figures

Related papers

View all related →