The central role of the loss function in reinforcement learning

DOI10.1214/25-STS989MaRDI QIDQ6860956zbMATH OpenFDO

Authors Kaiwen Wang, Nathan Kallus, Wen Sun

Publication date 10 February 2026

Published in Statistical Science (Search for Journal in Brave)

Full work available at URL https://projecteuclid.org/journals/statistical-science/volume-40/issue-4/The-Central-Role-of-the-Loss-Function-in-Reinforcement-Learning/10.1214/25-STS989.full

zbMATH Keywords

distributional RLfirst-order (small-loss) and second-order (variance-dependent) boundsRL with function approximation

Mathematics Subject Classification ID

Statistics (62-XX)

Cites work

• \({\mathcal Q}\)-learning
• Active learning for cost-sensitive classification
• An analysis of temporal-difference learning with function approximation
• Bridging Offline Reinforcement Learning and Imitation Learning: A Tale of Pessimism
• Combinatorial methods in density estimation
• Distributional Reinforcement Learning
• Exponential lower bounds for planning in MDPs with linearly-realizable optimal action-value functions
• Fast learning rates for plug-in classifiers
• Finite-time bounds for fitted value iteration
• High-dimensional statistics. A non-asymptotic viewpoint
• Mathematical Analysis of Machine Learning Algorithms
• The Nonstochastic Multiarmed Bandit Problem

This page was built for publication: The central role of the loss function in reinforcement learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6860956)