2 papers
math.PR2021
On the stability of the stochastic gradient Langevin algorithm with dependent data stream
Miklós Rásonyi, Kinga Tikosi
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 seque…
math.OC2020
Convergence of the Kiefer-Wolfowitz algorithm in the presence of discontinuities
Miklos Rasonyi, Kinga Tikosi
In this paper we estimate the expected error of a stochastic approximation algorithm where the maximum of a function is found using finite differences of a stochastic representatio…