Policy iteration based on stochastic factorization

DOI10.1613/JAIR.4301MaRDI QIDQ2878742zbMATH OpenOpenAlexWikidataFDO

Authors André M. S. Barreto, Joelle Pineau, Doina Precup

Publication date 5 September 2014

Published in The Journal of Artificial Intelligence Research (JAIR) (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1613/jair.4301

Mathematics Subject Classification ID

Dynamic programming (90C39) Markov and semi-Markov decision processes (90C40) Approximation methods and heuristics in mathematical programming (90C59)

Recommendations

• Factored value iteration converges
• Stochastic dynamic programming with factored representations
• Generalized polynomial approximations in Markovian decision processes
• State reduction in a Markov decision process
• The simplex and policy-iteration methods are strongly polynomial for the Markov decision problem with a fixed discount rate

Cited in

(8)

• Truncated policy iteration methods
• An incremental off-policy search in a model-free Markov decision process using a single sample path
• A numerical study of Markov decision process algorithms for multi-component replacement problems
• Factored value iteration converges
• 10.1162/1532443041827907
• scientific article; zbMATH DE number 2247690 (Why is no real title available?)
• Offline reinforcement learning in large state spaces: algorithms and guarantees
• The factored policy-gradient planner

This page was built for publication: Policy iteration based on stochastic factorization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2878742)