Markov Decision Processes Specified by Probabilistic Logic Programming: Representation and Solution

Thiago P. Bueno; Denis D. Maua; Leliane N. De Barros; Fabio G. Cozman

doi:10.1109/BRACIS.2016.068

Markov Decision Processes Specified by Probabilistic Logic Programming: Representation and Solution

Thiago P. Bueno, Denis D. Maua, Leliane N. De Barros, Fabio G. Cozman

Research output: Contribution to conference › Conference Paper

3 Scopus citations

Abstract

© 2016 IEEE. Probabilistic logic programming combines logic and probability, so as to obtain a rich modeling language. In this work, we extend ProbLog, a popular probabilistic logic programming language, with new constructs that allow the representation of (infinite-horizon) Markov decision processes. This new language can represent relational statements, including symmetric and transitive definitions, an advantage over other planning domain languages such as RDDL. We show how to exploit the logic structure in the language to perform Value Iteration. Preliminary experiments demonstrate the effectiveness of our framework.

Original language	American English
Pages	337-342
Number of pages	6
DOIs	https://doi.org/10.1109/BRACIS.2016.068
State	Published - 1 Feb 2017
Externally published	Yes
Event	Proceedings - 2016 5th Brazilian Conference on Intelligent Systems, BRACIS 2016 - Duration: 1 Feb 2017 → …

Conference

Conference	Proceedings - 2016 5th Brazilian Conference on Intelligent Systems, BRACIS 2016
Period	1/02/17 → …

Access to Document

10.1109/BRACIS.2016.068

Cite this

@conference{c081e64a6ee64c9da0aa32f2ef76fed3,

title = "Markov Decision Processes Specified by Probabilistic Logic Programming: Representation and Solution",

abstract = "{\textcopyright} 2016 IEEE. Probabilistic logic programming combines logic and probability, so as to obtain a rich modeling language. In this work, we extend ProbLog, a popular probabilistic logic programming language, with new constructs that allow the representation of (infinite-horizon) Markov decision processes. This new language can represent relational statements, including symmetric and transitive definitions, an advantage over other planning domain languages such as RDDL. We show how to exploit the logic structure in the language to perform Value Iteration. Preliminary experiments demonstrate the effectiveness of our framework.",

author = "Bueno, {Thiago P.} and Maua, {Denis D.} and {De Barros}, {Leliane N.} and Cozman, {Fabio G.}",

year = "2017",

month = feb,

day = "1",

doi = "10.1109/BRACIS.2016.068",

language = "American English",

pages = "337--342",

note = "Proceedings - 2016 5th Brazilian Conference on Intelligent Systems, BRACIS 2016 ; Conference date: 01-02-2017",

}

TY - CONF

T1 - Markov Decision Processes Specified by Probabilistic Logic Programming: Representation and Solution

AU - Bueno, Thiago P.

AU - Maua, Denis D.

AU - De Barros, Leliane N.

AU - Cozman, Fabio G.

PY - 2017/2/1

Y1 - 2017/2/1

N2 - © 2016 IEEE. Probabilistic logic programming combines logic and probability, so as to obtain a rich modeling language. In this work, we extend ProbLog, a popular probabilistic logic programming language, with new constructs that allow the representation of (infinite-horizon) Markov decision processes. This new language can represent relational statements, including symmetric and transitive definitions, an advantage over other planning domain languages such as RDDL. We show how to exploit the logic structure in the language to perform Value Iteration. Preliminary experiments demonstrate the effectiveness of our framework.

AB - © 2016 IEEE. Probabilistic logic programming combines logic and probability, so as to obtain a rich modeling language. In this work, we extend ProbLog, a popular probabilistic logic programming language, with new constructs that allow the representation of (infinite-horizon) Markov decision processes. This new language can represent relational statements, including symmetric and transitive definitions, an advantage over other planning domain languages such as RDDL. We show how to exploit the logic structure in the language to perform Value Iteration. Preliminary experiments demonstrate the effectiveness of our framework.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85015183645&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=85015183645&origin=inward

U2 - 10.1109/BRACIS.2016.068

DO - 10.1109/BRACIS.2016.068

M3 - Conference Paper

SP - 337

EP - 342

T2 - Proceedings - 2016 5th Brazilian Conference on Intelligent Systems, BRACIS 2016

Y2 - 1 February 2017

ER -

Markov Decision Processes Specified by Probabilistic Logic Programming: Representation and Solution

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this