English

Quadratic number of nodes is sufficient to learn a dataset via gradient descent

Optimization and Control 2019-11-14 v1 Machine Learning Statistics Theory Statistics Theory

Abstract

We prove that if an activation function satisfies some mild conditions and number of neurons in a two-layered fully connected neural network with this activation function is beyond a certain threshold, then gradient descent on quadratic loss function finds the optimal weights of input layer for global minima in linear time. This threshold value is an improvement over previously obtained values. We hypothesise that this bound cannot be improved by the method we are using in this work.

Keywords

Cite

@article{arxiv.1911.05402,
  title  = {Quadratic number of nodes is sufficient to learn a dataset via gradient descent},
  author = {Biswarup Das and Eugene. A. Golikov},
  journal= {arXiv preprint arXiv:1911.05402},
  year   = {2019}
}

Comments

Machine learning using neural networks, gradient descent, optimization, overparametrization regime