Saved in:
Bibliographic Details
Main Authors: Nguyen-Tang, Thanh, Arora, Raman
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.06339
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • We study the statistical complexity of offline decision-making with function approximation, establishing (near) minimax-optimal rates for stochastic contextual bandits and Markov decision processes. The performance limits are captured by the pseudo-dimension of the (value) function class and a new characterization of the behavior policy that \emph{strictly} subsumes all the previous notions of data coverage in the offline decision-making literature. In addition, we seek to understand the benefits of using offline data in online decision-making and show nearly minimax-optimal rates in a wide range of regimes.