Gap-Dependent Unsupervised Exploration for Reinforcement Learning

Code

• github.com/uuujf/GapExploration
  Officialnone★ 0

Abstract

For the problem of task-agnostic reinforcement learning (RL), an agent first collects samples from an unknown environment without the supervision of reward signals, then is revealed with a reward and is asked to compute a corresponding near-optimal policy. Existing approaches mainly concern the worst-case scenarios, in which no structural information of the reward/transition-dynamics is utilized. Therefore the best sample upper bound is O(1/^2), where >0 is the target accuracy of the obtained policy, and can be overly pessimistic. To tackle this issue, we provide an efficient algorithm that utilizes a gap parameter, >0, to reduce the amount of exploration. In particular, for an unknown finite-horizon Markov decision process, the algorithm takes only O (1/ (H^3SA / + H^4 S^2 A) ) episodes of exploration, and is able to obtain an -optimal policy for a post-revealed reward with sub-optimality gap at least , where S is the number of states, A is the number of actions, and H is the length of the horizon, obtaining a nearly quadratic saving in terms of . We show that, information-theoretically, this bound is nearly tight for < (1/(HS)) and H>1. We further show that O(1) sample bound is possible for H=1 (i.e., multi-armed bandit) or with a sampling simulator, establishing a stark separation between those settings and the RL setting.

Tasks

Reproductions