Singular perturbations in stochastic optimal control with unbounded data
Optimization and Control
2023-03-29 v2 Analysis of PDEs
Probability
Abstract
We study singular perturbations of a class of two-scale stochastic control systems with unbounded data. The assumptions are designed to cover some relaxation problems for deep neural networks. We construct effective Hamiltonian and initial data and prove the convergence of the value function to the solution of a limit (effective) Cauchy problem for a parabolic equation of HJB type. We use methods of probability, viscosity solutions and homogenization.
Cite
@article{arxiv.2208.00655,
title = {Singular perturbations in stochastic optimal control with unbounded data},
author = {Martino Bardi and Hicham Kouhkouh},
journal= {arXiv preprint arXiv:2208.00655},
year = {2023}
}
Comments
26 pages