Gradient-free two-points optimal method for non smooth stochastic convex optimization problem with additional small noise
Optimization and Control
2017-08-15 v4
Abstract
Using double-smoothing technique and stochastic mirror descent with inexact oracle we built an optimal algorithm (up to a multiplicative factor) for two-points gradient-free non-smooth stochastic convex programming. We investigate how much can be the level of noise (the nature of this noise isn't necessary stochastic) for the rate of convergence to be maintained (up to a multiplicative factor).
Cite
@article{arxiv.1701.03821,
title = {Gradient-free two-points optimal method for non smooth stochastic convex optimization problem with additional small noise},
author = {Anastasia Bayandina and Alexander Gasnikov and Fariman Guliev and Anastasia Lagunovskaya},
journal= {arXiv preprint arXiv:1701.03821},
year = {2017}
}
Comments
in Russian, 13 pages; Automatic and Remote Control, 2018