Heuristic variable grid solution method for POMDPs

Ronen I. Brafman

Heuristic variable grid solution method for POMDPs

Research output: Contribution to conference › Paper › peer-review

38 Scopus citations

Abstract

Partially observable Markov decision processes (POMDPs) are an appealing tool for modeling planning problems under uncertainty. They incorporate stochastic action and sensor descriptions and easily capture goal oriented and process oriented tasks. Unfortunately, POMDPs are very difficult to solve. Exact methods cannot handle problems with much more than 10 states, so approximate methods must be used. In this paper, we describe a simple variable-grid solution method which yields good results on relatively large problems with modest computational effort.

Original language	English
Pages	727-733
Number of pages	7
State	Published - 1 Dec 1997
Externally published	Yes
Event	Proceedings of the 1997 14th National Conference on Artificial Intelligence, AAAI 97 - Providence, RI, USA Duration: 27 Jul 1997 → 31 Jul 1997

Conference

Conference	Proceedings of the 1997 14th National Conference on Artificial Intelligence, AAAI 97
City	Providence, RI, USA
Period	27/07/97 → 31/07/97

ASJC Scopus subject areas

Software
Artificial Intelligence

Cite this

@conference{26339937cabc49278205390b9cbca0d3,

title = "Heuristic variable grid solution method for POMDPs",

abstract = "Partially observable Markov decision processes (POMDPs) are an appealing tool for modeling planning problems under uncertainty. They incorporate stochastic action and sensor descriptions and easily capture goal oriented and process oriented tasks. Unfortunately, POMDPs are very difficult to solve. Exact methods cannot handle problems with much more than 10 states, so approximate methods must be used. In this paper, we describe a simple variable-grid solution method which yields good results on relatively large problems with modest computational effort.",

author = "Brafman, {Ronen I.}",

year = "1997",

month = dec,

day = "1",

language = "English",

pages = "727--733",

note = "Proceedings of the 1997 14th National Conference on Artificial Intelligence, AAAI 97 ; Conference date: 27-07-1997 Through 31-07-1997",

}

TY - CONF

T1 - Heuristic variable grid solution method for POMDPs

AU - Brafman, Ronen I.

PY - 1997/12/1

Y1 - 1997/12/1

N2 - Partially observable Markov decision processes (POMDPs) are an appealing tool for modeling planning problems under uncertainty. They incorporate stochastic action and sensor descriptions and easily capture goal oriented and process oriented tasks. Unfortunately, POMDPs are very difficult to solve. Exact methods cannot handle problems with much more than 10 states, so approximate methods must be used. In this paper, we describe a simple variable-grid solution method which yields good results on relatively large problems with modest computational effort.

AB - Partially observable Markov decision processes (POMDPs) are an appealing tool for modeling planning problems under uncertainty. They incorporate stochastic action and sensor descriptions and easily capture goal oriented and process oriented tasks. Unfortunately, POMDPs are very difficult to solve. Exact methods cannot handle problems with much more than 10 states, so approximate methods must be used. In this paper, we describe a simple variable-grid solution method which yields good results on relatively large problems with modest computational effort.

UR - http://www.scopus.com/inward/record.url?scp=0031385391&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0031385391

SP - 727

EP - 733

T2 - Proceedings of the 1997 14th National Conference on Artificial Intelligence, AAAI 97

Y2 - 27 July 1997 through 31 July 1997

ER -

Heuristic variable grid solution method for POMDPs

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this