<ul><li>Nonlinear control systems with partial information to the decision maker are prevalent in a variety of applications.</li><li>This work explores reinforcement learning methods for finding the optimal policy in the nearly linear-quadratic regulator systems.</li><li>The cost function of the system is nonconvex, but the study establishes local strong convexity and smoothness in the vicinity of the global optimizer.</li><li>A policy gradient algorithm is proposed that is guaranteed to converge to the globally optimal policy with a linear rate.</li></ul>

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

Discover more