On the policy improvement algorithm for ergodic risk-sensitive control
Optimization and Control
2021-01-01 v3 Probability
Abstract
In this article we consider the ergodic risk-sensitive control problem for a large class of multidimensional controlled diffusions on the whole space. We study the minimization and maximization problems under either a blanket stability hypothesis, or a near-monotone assumption on the running cost. We establish the convergence of the policy improvement algorithm for these models. We also present a more general result concerning the region of attraction of the equilibrium of the algorithm.
Keywords
Cite
@article{arxiv.1912.04568,
title = {On the policy improvement algorithm for ergodic risk-sensitive control},
author = {Ari Arapostathis and Anup Biswas and Somnath Pradhan},
journal= {arXiv preprint arXiv:1912.04568},
year = {2021}
}
Comments
20 pages