A Sample-Efficient Algorithm for Episodic Finite-Horizon MDP with Constraints

09/23/2020 ∙ by Krishna C. Kalagarla, et al. ∙ 3

Constrained Markov Decision Processes (CMDPs) formalize sequential decision-making problems whose objective is to minimize a cost function while satisfying constraints on various cost functions. In this paper, we consider the setting of episodic fixed-horizon CMDPs. We propose an online algorithm which leverages the linear programming formulation of finite-horizon CMDP for repeated optimistic planning to provide a probably approximately correct (PAC) guarantee on the number of episodes needed to ensure an ϵ-optimal policy, i.e., with resulting objective value within ϵ of the optimal value and satisfying the constraints within ϵ-tolerance, with probability at least 1-δ. The number of episodes needed is shown to be of the order 𝒪̃(|S||A|C^2H^2/ϵ^2log1/δ), where C is the upper bound on the number of possible successor states for a state-action pair. Therefore, if C ≪ |S|, the number of episodes needed have a linear dependence on the state and action space sizes |S| and |A|, respectively, and quadratic dependence on the time horizon H.



There are no comments yet.


page 1

page 2

page 3

page 4

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.