An efficient gradient projection method for stochastic optimal control problem with expected integral state constraint

TL;DR

This paper introduces an efficient gradient projection method for solving stochastic optimal control problems with expected integral state constraints, combining discretization, Monte Carlo approximation, and error analysis.

Contribution

It develops a novel gradient projection approach with linear drift and specific multipliers for constrained stochastic control, including convergence analysis and numerical validation.

Findings

Method achieves first-order convergence.

Numerical examples confirm theoretical results.

Efficient handling of state constraints in stochastic control.

Abstract

In this work, we present an efficient gradient projection method for solving a class of stochastic optimal control problem with expected integral state constraint. The first order optimality condition system consisting of forward-backward stochastic differential equations and a variational equation is first derived. Then, an efficient gradient projection method with linear drift coefficient is proposed where the state constraint is guaranteed by constructing specific multiplier. Further, the Euler method is used to discretize the forward-backward stochastic differential equations and the associated conditional expectations are approximated by the least square Monte Carlo method, yielding the fully discrete iterative scheme. Error estimates of control and multiplier are presented, showing that the method admits first order convergence. Finally we present numerical examples to support the theoretical findings.