Сохранить в:
Библиографические подробности
Главные авторы: Nguyen-Tang, Thanh, Arora, Raman
Формат: Preprint
Опубликовано: 2025
Предметы:
Online-ссылка:https://arxiv.org/abs/2501.06339
Метки: Добавить метку
Нет меток, Требуется 1-ая метка записи!
_version_ 1866909454376632320
author Nguyen-Tang, Thanh
Arora, Raman
author_facet Nguyen-Tang, Thanh
Arora, Raman
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.
format Preprint
id arxiv_https___arxiv_org_abs_2501_06339
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On The Statistical Complexity of Offline Decision-Making
Nguyen-Tang, Thanh
Arora, Raman
Machine Learning
Artificial Intelligence
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.
title On The Statistical Complexity of Offline Decision-Making
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2501.06339