Distributionally robust stochastic optimal control

TL;DR

This paper explores the development of distributionally robust formulations of stochastic optimal control problems, analyzing conditions for the existence of non-randomized policies to enhance decision-making under uncertainty.

Contribution

It provides necessary and sufficient conditions for the existence of non-randomized policies in distributionally robust stochastic control, advancing theoretical understanding.

Findings

Conditions for non-randomized policy existence established

Framework for distributionally robust control formulated

Theoretical insights into policy robustness obtained

Abstract

The main goal of this paper is to discuss the construction of distributionally robust counterparts of stochastic optimal control problems. Randomized and non-randomized policies are considered. In particular, necessary and sufficient conditions for the existence of non-randomized policies are given.