English

Strongly convex stochastic online optimization on a unit simplex with application to the mixing least square regression

Optimization and Control 2017-03-24 v2

Abstract

In this paper we propose a new approach to obtain mixing least square regression estimate by means of stochastic online mirror descent in non-euclidian set-up.

Keywords

Cite

@article{arxiv.1703.06770,
  title  = {Strongly convex stochastic online optimization on a unit simplex with application to the mixing least square regression},
  author = {Anastasia Bayandina and Elena Chernousova and Alexander Gasnikov and Ekaterina Krymova},
  journal= {arXiv preprint arXiv:1703.06770},
  year   = {2017}
}

Comments

This paper has been withdrawn by the author due to a crucial error in Theorem 1 and in item 3