all AI news
On the Statistical Efficiency of Reward-Free Exploration in Non-Linear RL. (arXiv:2206.10770v1 [cs.LG])
cs.LG updates on arXiv.org arxiv.org
We study reward-free reinforcement learning (RL) under general non-linear
function approximation, and establish sample efficiency and hardness results
under various standard structural assumptions. On the positive side, we propose
the RFOLIVE (Reward-Free OLIVE) algorithm for sample-efficient reward-free
exploration under minimal structural assumptions, which covers the previously
studied settings of linear MDPs (Jin et al., 2020b), linear completeness
(Zanette et al., 2020b) and low-rank MDPs with unknown representation (Modi et
al., 2021). Our analyses indicate that the explorability or reachability
assumptions, …
arxiv efficiency exploration free lg linear non-linear rl statistical