Convergence of the Kiefer–Wolfowitz algorithm in the presence of discontinuities

Rásonyi, Miklós and Tikosi, Kinga (2023) Convergence of the Kiefer–Wolfowitz algorithm in the presence of discontinuities. ADVANCES IN APPLIED PROBABILITY, 55 (2). pp. 382-406. ISSN 0001-8678

Preview

Text 2007.14069v2.pdf Available under License Creative Commons Attribution. Download (668kB) | Preview

Abstract

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 representation of that function. An error estimate of the order for the n th iteration is achieved using suitable parameters. The novelty with respect to previous studies is that we allow the stochastic representation to be discontinuous and to consist of possibly dependent random variables (satisfying a mixing condition).

Actions (login required)

Edit Item