English

Analysis of the rate of convergence of an over-parametrized convolutional neural network image classifier learned by gradient descent

Machine Learning 2024-05-14 v1 Machine Learning

Abstract

Image classification based on over-parametrized convolutional neural networks with a global average-pooling layer is considered. The weights of the network are learned by gradient descent. A bound on the rate of convergence of the difference between the misclassification risk of the newly introduced convolutional neural network estimate and the minimal possible value is derived.

Keywords

Cite

@article{arxiv.2405.07619,
  title  = {Analysis of the rate of convergence of an over-parametrized convolutional neural network image classifier learned by gradient descent},
  author = {Michael Kohler and Adam Krzyzak and Benjamin Walter},
  journal= {arXiv preprint arXiv:2405.07619},
  year   = {2024}
}