Efficient Planning in Large MDPs with Weak Linear Function Approximation

Abstract

Large-scale Markov decision processes (MDPs) require planning algorithms with runtime independent of the number of states of the MDP. We consider the planning problem in MDPs using linear value function approximation with only weak requirements: low approximation error for the optimal value function, and a small set of "core" states whose features span those of other states. In particular, we make no assumptions about the representability of policies or value functions of non-optimal policies. Our algorithm produces almost-optimal actions for any state using a generative oracle (simulator) for the MDP, while its computation time scales polynomially with the number of features, core states, and actions and the effective horizon.

12 pages and appendix (10 pages). Submitted to the 34th Conference on Neural Information Processing Systems (NeurIPS 2020), Vancouver, Canada

References in corpus (5)

• Contextual Decision Processes with Low Bellman Rank are PAC-Learnable
• Stochastic Primal-Dual Methods and Sample Complexity of Reinforcement Learning
• Is a Good Representation Sufficient for Sample Efficient Reinforcement Learning?
• Comments on the Du-Kakade-Wang-Yang Lower Bounds
• Large-Scale Markov Decision Problems via the Linear Programming Dual