On the stability of the stochastic gradient Langevin algorithm with dependent data stream
Probability
2021-05-05 v1 Optimization and Control
Statistics Theory
Statistics Theory
Abstract
We prove, under mild conditions, that the stochastic gradient Langevin dynamics converges to a limiting law as time tends to infinity, even in the case where the driving data sequence is dependent.
Keywords
Cite
@article{arxiv.2105.01422,
title = {On the stability of the stochastic gradient Langevin algorithm with dependent data stream},
author = {Miklós Rásonyi and Kinga Tikosi},
journal= {arXiv preprint arXiv:2105.01422},
year = {2021}
}