The infinite regionalized policy representation

Miao Liu; Xuejun Liao; Lawrence Carin

The infinite regionalized policy representation

Miao Liu, Xuejun Liao, Lawrence Carin

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We introduce the infinite regionalized policy presentation (iRPR), as a nonparametric policy for reinforcement learning in partially observable Markov decision processes (POMDPs). The iRPR assumes an unbounded set of decision states a priori, and infers the number of states to represent the policy given the experiences. We propose algorithms for learning the number of decision states while maintaining a proper balance between exploration and exploitation. Convergence analysis is provided, along with performance evaluations on benchmark problems. Copyright 2011 by the author(s)/owner(s).

Original language	English (US)
Title of host publication	Proceedings of the 28th International Conference on Machine Learning, ICML 2011
Pages	769-776
Number of pages	8
State	Published - Oct 7 2011
Externally published	Yes

Bibliographical note

Generated from Scopus record by KAUST IRTS on 2021-02-09

Cite this

@inproceedings{469de398797042b5bfea16307f7b89df,

title = "The infinite regionalized policy representation",

abstract = "We introduce the infinite regionalized policy presentation (iRPR), as a nonparametric policy for reinforcement learning in partially observable Markov decision processes (POMDPs). The iRPR assumes an unbounded set of decision states a priori, and infers the number of states to represent the policy given the experiences. We propose algorithms for learning the number of decision states while maintaining a proper balance between exploration and exploitation. Convergence analysis is provided, along with performance evaluations on benchmark problems. Copyright 2011 by the author(s)/owner(s).",

author = "Miao Liu and Xuejun Liao and Lawrence Carin",

note = "Generated from Scopus record by KAUST IRTS on 2021-02-09",

year = "2011",

month = oct,

day = "7",

language = "English (US)",

isbn = "9781450306195",

pages = "769--776",

booktitle = "Proceedings of the 28th International Conference on Machine Learning, ICML 2011",

}

TY - GEN

T1 - The infinite regionalized policy representation

AU - Liu, Miao

AU - Liao, Xuejun

AU - Carin, Lawrence

N1 - Generated from Scopus record by KAUST IRTS on 2021-02-09

PY - 2011/10/7

Y1 - 2011/10/7

N2 - We introduce the infinite regionalized policy presentation (iRPR), as a nonparametric policy for reinforcement learning in partially observable Markov decision processes (POMDPs). The iRPR assumes an unbounded set of decision states a priori, and infers the number of states to represent the policy given the experiences. We propose algorithms for learning the number of decision states while maintaining a proper balance between exploration and exploitation. Convergence analysis is provided, along with performance evaluations on benchmark problems. Copyright 2011 by the author(s)/owner(s).

AB - We introduce the infinite regionalized policy presentation (iRPR), as a nonparametric policy for reinforcement learning in partially observable Markov decision processes (POMDPs). The iRPR assumes an unbounded set of decision states a priori, and infers the number of states to represent the policy given the experiences. We propose algorithms for learning the number of decision states while maintaining a proper balance between exploration and exploitation. Convergence analysis is provided, along with performance evaluations on benchmark problems. Copyright 2011 by the author(s)/owner(s).

UR - http://www.scopus.com/inward/record.url?scp=80053453875&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781450306195

SP - 769

EP - 776

BT - Proceedings of the 28th International Conference on Machine Learning, ICML 2011

ER -

The infinite regionalized policy representation

Abstract

Bibliographical note

Other files and links

Fingerprint

Cite this