Analysis of classification-based policy iteration algorithms

MaRDI QIDQ2810787zbMATH OpenFDO

Authors Alessandro Lazaric, Mohammad Ghavamzadeh, Rémi Munos

Publication date 6 June 2016

Published in Journal of Machine Learning Research (JMLR) (Search for Journal in Brave)

Full work available at URL http://jmlr.csail.mit.edu/papers/v17/10-364.html

zbMATH Keywords

classification-based approach to policy iterationfinite-sample analysispolicy iterationreinforcement learning

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Markov and semi-Markov decision processes (90C40)

Recommendations

• Rollout sampling approximate policy iteration
• Finite-sample analysis of least-squares policy iteration
• Approximate policy iteration: a survey and some new methods
• Regularized policy iteration with nonparametric function spaces
• Performance Loss Bounds for Approximate Value Iteration with State Aggregation

Cited in

(8)

• Rollout sampling approximate policy iteration
• Performance guarantees for policy learning
• Preference-based reinforcement learning: evolutionary direct policy search using a preference-based racing algorithm
• Safe policy iteration: a monotonically improving approximate policy iteration approach
• On the theory of policy gradient methods: optimality, approximation, and distribution shift
• A priori estimates for deep residual network in continuous-time reinforcement learning
• Deep approximate policy iteration
• Deep controlled learning for inventory control

This page was built for publication: Analysis of classification-based policy iteration algorithms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2810787)