Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty

Peeyush Kumar; Archis Ghate

doi:10.1007/978-3-030-04726-9_20

Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty

Peeyush Kumar
, Archis Ghate

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

Abstract

This paper formulates partially observable Markov decision processes, where state-transition probabilities and measurement outcome probabilities are characterized by unknown parameters. An information theoretic solution method that adaptively manages the resulting exploitation-exploration trade-off is proposed. Numerical experiments for response guided dosing in healthcare are presented.

Original language	English (US)
Title of host publication	Advances in Service Science - Proceedings of the 2018 INFORMS International Conference on Service Science
Editors	Hui Yang, Robin Qiu
Publisher	Springer Science and Business Media B.V.
Pages	201-209
Number of pages	9
ISBN (Print)	9783030047252
DOIs	https://doi.org/10.1007/978-3-030-04726-9_20
State	Published - 2019
Externally published	Yes
Event	INFORMS International Conference on Service Science, ICSS 2018 - Phoenix, United States Duration: Nov 3 2018 → Nov 3 2018

Publication series

Name	Springer Proceedings in Business and Economics
ISSN (Print)	2198-7246
ISSN (Electronic)	2198-7254

Conference

Conference	INFORMS International Conference on Service Science, ICSS 2018
Country/Territory	United States
City	Phoenix
Period	11/3/18 → 11/3/18

Bibliographical note

Publisher Copyright:
© 2019, Springer Nature Switzerland AG.

Publisher link

10.1007/978-3-030-04726-9_20

Find It

Find It at U of M Libraries

Cite this

Kumar, P., & Ghate, A. (2019). Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty. In H. Yang, & R. Qiu (Eds.), Advances in Service Science - Proceedings of the 2018 INFORMS International Conference on Service Science (pp. 201-209). (Springer Proceedings in Business and Economics). Springer Science and Business Media B.V.. https://doi.org/10.1007/978-3-030-04726-9_20

Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty. / Kumar, Peeyush; Ghate, Archis.
Advances in Service Science - Proceedings of the 2018 INFORMS International Conference on Service Science. ed. / Hui Yang; Robin Qiu. Springer Science and Business Media B.V., 2019. p. 201-209 (Springer Proceedings in Business and Economics).

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

Kumar, P & Ghate, A 2019, Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty. in H Yang & R Qiu (eds), Advances in Service Science - Proceedings of the 2018 INFORMS International Conference on Service Science. Springer Proceedings in Business and Economics, Springer Science and Business Media B.V., pp. 201-209, INFORMS International Conference on Service Science, ICSS 2018, Phoenix, United States, 11/3/18. https://doi.org/10.1007/978-3-030-04726-9_20

Kumar P, Ghate A. Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty. In Yang H, Qiu R, editors, Advances in Service Science - Proceedings of the 2018 INFORMS International Conference on Service Science. Springer Science and Business Media B.V. 2019. p. 201-209. (Springer Proceedings in Business and Economics). doi: 10.1007/978-3-030-04726-9_20

Kumar, Peeyush ; Ghate, Archis. / Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty. Advances in Service Science - Proceedings of the 2018 INFORMS International Conference on Service Science. editor / Hui Yang ; Robin Qiu. Springer Science and Business Media B.V., 2019. pp. 201-209 (Springer Proceedings in Business and Economics).

@inproceedings{0d7573d7af404fb084179f408ccffa7a,

title = "Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty",

abstract = "This paper formulates partially observable Markov decision processes, where state-transition probabilities and measurement outcome probabilities are characterized by unknown parameters. An information theoretic solution method that adaptively manages the resulting exploitation-exploration trade-off is proposed. Numerical experiments for response guided dosing in healthcare are presented.",

author = "Peeyush Kumar and Archis Ghate",

note = "Publisher Copyright: {\textcopyright} 2019, Springer Nature Switzerland AG.; INFORMS International Conference on Service Science, ICSS 2018 ; Conference date: 03-11-2018 Through 03-11-2018",

year = "2019",

doi = "10.1007/978-3-030-04726-9\_20",

language = "English (US)",

isbn = "9783030047252",

series = "Springer Proceedings in Business and Economics",

publisher = "Springer Science and Business Media B.V.",

pages = "201--209",

editor = "Hui Yang and Robin Qiu",

booktitle = "Advances in Service Science - Proceedings of the 2018 INFORMS International Conference on Service Science",

}

TY - GEN

T1 - Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty

AU - Kumar, Peeyush

AU - Ghate, Archis

PY - 2019

Y1 - 2019

N2 - This paper formulates partially observable Markov decision processes, where state-transition probabilities and measurement outcome probabilities are characterized by unknown parameters. An information theoretic solution method that adaptively manages the resulting exploitation-exploration trade-off is proposed. Numerical experiments for response guided dosing in healthcare are presented.

AB - This paper formulates partially observable Markov decision processes, where state-transition probabilities and measurement outcome probabilities are characterized by unknown parameters. An information theoretic solution method that adaptively manages the resulting exploitation-exploration trade-off is proposed. Numerical experiments for response guided dosing in healthcare are presented.

UR - https://www.scopus.com/pages/publications/85126198910

UR - https://www.scopus.com/pages/publications/85126198910#tab=citedBy

U2 - 10.1007/978-3-030-04726-9_20

DO - 10.1007/978-3-030-04726-9_20

M3 - Conference contribution

AN - SCOPUS:85126198910

SN - 9783030047252

T3 - Springer Proceedings in Business and Economics

SP - 201

EP - 209

BT - Advances in Service Science - Proceedings of the 2018 INFORMS International Conference on Service Science

A2 - Yang, Hui

A2 - Qiu, Robin

PB - Springer Science and Business Media B.V.

T2 - INFORMS International Conference on Service Science, ICSS 2018

Y2 - 3 November 2018 through 3 November 2018

ER -

Information Directed Policy Sampling for Partially Observable Markov Decision Processes with Parametric Uncertainty

Abstract

Publication series

Conference

Bibliographical note

Publisher link

Find It

Other files and links

Fingerprint

Cite this