Reinforcement learning with limited reinforcement: using Bayes risk for active learning in POMDPs

DOI10.1016/j.artint.2012.04.006zbMath1251.68177OpenAlexW1973749650MaRDI QIDQ1761294

Nicholas Roy, Joelle Pineau, Finale Doshi-Velez

Publication date: 15 November 2012

Published in: Artificial Intelligence (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.artint.2012.04.006

zbMATH Keywords

reinforcement learningBayesian methodspartially observable Markov decision process

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Related Items (3)

POMDP controllers with optimal budget ⋮ Reinforcement learning with limited reinforcement: using Bayes risk for active learning in POMDPs ⋮ A Bayesian learning model for estimating unknown demand parameter in revenue management

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Partially observable Markov decision processes with imprecise parameters
• Reinforcement learning with limited reinforcement: using Bayes risk for active learning in POMDPs
• A Partially Observed Markov Decision Process for Dynamic Pricing
• Sequential Monte Carlo Samplers
• Workspace-Based Connectivity Oracle: An Adaptive Sampling Strategy for PRM Planning
• The Optimal Control of Partially Observable Markov Processes over the Infinite Horizon: Discounted Costs
• Bayesian Methods for Hidden Markov Models
• Robust Control of Markov Decision Processes with Uncertain Transition Matrices

This page was built for publication: Reinforcement learning with limited reinforcement: using Bayes risk for active learning in POMDPs