Policy Gradient Converges to the Globally Optimal Policy for Nearly Linear-Quadratic Regulators

arXiv:2303.08431v3 Announce Type: replace
Abstract: Nonlinear control systems with partial information to the decision maker are prevalent in a variety of applications. As a step toward studying such nonlinear systems, this work explores reinforcement learning methods for finding the optimal policy in the nearly linear-quadratic regulator systems. In particular, we consider a dynamic system that combines linear and nonlinear components, and is governed by a policy with the same structure. Assuming that the nonlinear component comprises kernels with small Lipschitz …

abstract applications arxiv control control systems cs.lg decision gradient information linear maker math.oc policy regulator regulators reinforcement reinforcement learning stat.ml studying systems type work