English

Nearly Minimax Optimal Reinforcement Learning for Linear Mixture Markov Decision Processes

Machine Learning 2021-01-08 v2 Optimization and Control Machine Learning

Abstract

We study reinforcement learning (RL) with linear function approximation where the underlying transition probability kernel of the Markov decision process (MDP) is a linear mixture model (Jia et al., 2020; Ayoub et al., 2020; Zhou et al., 2020) and the learning agent has access to either an integration or a sampling oracle of the individual basis kernels. We propose a new Bernstein-type concentration inequality for self-normalized martingales for linear bandit problems with bounded noise. Based on the new inequality, we propose a new, computationally efficient algorithm with linear function approximation named UCRL-VTR+\text{UCRL-VTR}^{+} for the aforementioned linear mixture MDPs in the episodic undiscounted setting. We show that UCRL-VTR+\text{UCRL-VTR}^{+} attains an O~(dHT)\tilde O(dH\sqrt{T}) regret where dd is the dimension of feature mapping, HH is the length of the episode and TT is the number of interactions with the MDP. We also prove a matching lower bound Ω(dHT)\Omega(dH\sqrt{T}) for this setting, which shows that UCRL-VTR+\text{UCRL-VTR}^{+} is minimax optimal up to logarithmic factors. In addition, we propose the UCLK+\text{UCLK}^{+} algorithm for the same family of MDPs under discounting and show that it attains an O~(dT/(1γ)1.5)\tilde O(d\sqrt{T}/(1-\gamma)^{1.5}) regret, where γ[0,1)\gamma\in [0,1) is the discount factor. Our upper bound matches the lower bound Ω(dT/(1γ)1.5)\Omega(d\sqrt{T}/(1-\gamma)^{1.5}) proved by Zhou et al. (2020) up to logarithmic factors, suggesting that UCLK+\text{UCLK}^{+} is nearly minimax optimal. To the best of our knowledge, these are the first computationally efficient, nearly minimax optimal algorithms for RL with linear function approximation.

Keywords

Cite

@article{arxiv.2012.08507,
  title  = {Nearly Minimax Optimal Reinforcement Learning for Linear Mixture Markov Decision Processes},
  author = {Dongruo Zhou and Quanquan Gu and Csaba Szepesvari},
  journal= {arXiv preprint arXiv:2012.08507},
  year   = {2021}
}

Comments

59 pages, 1 figure

R2 v1 2026-06-23T20:59:42.103Z