Singular Value Decomposition and Neural Networks
Machine Learning
2019-09-16 v1 Numerical Analysis
Numerical Analysis
Machine Learning
Abstract
Singular Value Decomposition (SVD) constitutes a bridge between the linear algebra concepts and multi-layer neural networks---it is their linear analogy. Besides of this insight, it can be used as a good initial guess for the network parameters, leading to substantially better optimization results.
Cite
@article{arxiv.1906.11755,
title = {Singular Value Decomposition and Neural Networks},
author = {Bernhard Bermeitinger and Tomas Hrycej and Siegfried Handschuh},
journal= {arXiv preprint arXiv:1906.11755},
year = {2019}
}