53
11

Geometric Exploration for Online Control

Abstract

We study the control of an \emph{unknown} linear dynamical system under general convex costs. The objective is minimizing regret vs. the class of disturbance-feedback-controllers, which encompasses all stabilizing linear-dynamical-controllers. In this work, we first consider the case of known cost functions, for which we design the first polynomial-time algorithm with n3Tn^3\sqrt{T}-regret, where nn is the dimension of the state plus the dimension of control input. The T\sqrt{T}-horizon dependence is optimal, and improves upon the previous best known bound of T2/3T^{2/3}. The main component of our algorithm is a novel geometric exploration strategy: we adaptively construct a sequence of barycentric spanners in the policy space. Second, we consider the case of bandit feedback, for which we give the first polynomial-time algorithm with poly(n)Tpoly(n)\sqrt{T}-regret, building on Stochastic Bandit Convex Optimization.

View on arXiv
Comments on this paper