English

Finite Sample Analysis of Minimax Offline Reinforcement Learning: Completeness, Fast Rates and First-Order Efficiency

Machine Learning 2022-07-26 v2 Statistics Theory Machine Learning Statistics Theory

Abstract

We offer a theoretical characterization of off-policy evaluation (OPE) in reinforcement learning using function approximation for marginal importance weights and qq-functions when these are estimated using recent minimax methods. Under various combinations of realizability and completeness assumptions, we show that the minimax approach enables us to achieve a fast rate of convergence for weights and quality functions, characterized by the critical inequality \citep{bartlett2005}. Based on this result, we analyze convergence rates for OPE. In particular, we introduce novel alternative completeness conditions under which OPE is feasible and we present the first finite-sample result with first-order efficiency in non-tabular environments, i.e., having the minimal coefficient in the leading term.

Keywords

Cite

@article{arxiv.2102.02981,
  title  = {Finite Sample Analysis of Minimax Offline Reinforcement Learning: Completeness, Fast Rates and First-Order Efficiency},
  author = {Masatoshi Uehara and Masaaki Imaizumi and Nan Jiang and Nathan Kallus and Wen Sun and Tengyang Xie},
  journal= {arXiv preprint arXiv:2102.02981},
  year   = {2022}
}

Comments

Under Review