Average performance analysis of the stochastic gradient method for online PCA
Statistics Theory
2018-04-04 v1 Machine Learning
Machine Learning
Statistics Theory
Abstract
This paper studies the complexity of the stochastic gradient algorithm for PCA when the data are observed in a streaming setting. We also propose an online approach for selecting the learning rate. Simulation experiments confirm the practical relevance of the plain stochastic gradient approach and that drastic improvements can be achieved by learning the learning rate.
Cite
@article{arxiv.1804.01071,
title = {Average performance analysis of the stochastic gradient method for online PCA},
author = {Stephane Chretien and Christophe Guyeux and Zhen-Wai Olivier HO},
journal= {arXiv preprint arXiv:1804.01071},
year = {2018}
}
Comments
11 pages, 1 figure, Submitted to LOD 2018