On the overfly algorithm in deep learning of neural networks
Machine Learning
2019-01-15 v6 Dynamical Systems
Machine Learning
Abstract
In this paper we investigate the supervised backpropagation training of multilayer neural networks from a dynamical systems point of view. We discuss some links with the qualitative theory of differential equations and introduce the overfly algorithm to tackle the local minima problem. Our approach is based on the existence of first integrals of the generalised gradient system with build-in dissipation.
Cite
@article{arxiv.1807.10668,
title = {On the overfly algorithm in deep learning of neural networks},
author = {Alexei Tsygvintsev},
journal= {arXiv preprint arXiv:1807.10668},
year = {2019}
}
Comments
To appear in Applied Mathematics and Computation