The Theory from Large Deviations for Random Processes and Strong Convergence of Stochastic Approximation Procedures

Leonov, S. (1986). The Theory from Large Deviations for Random Processes and Strong Convergence of Stochastic Approximation Procedures. IIASA Working Paper. IIASA, Laxenburg, Austria: WP-86-057

Preview

Text WP-86-057.pdf Download (589kB) | Preview

Abstract

This paper deals with the application of "large deviation" theory to the analysis of stochastic approximation procedures. The approach allows to get new results in the asymptotical behaviour of stochastic procedures under very mild assumption about the "noise". The paper contains a short but illuminative survey of these results together with some new author's findings. For applications the last section seems to be interesting presenting some new ideas in multiobjective optimization.

Actions (login required)

View Item