Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability

José Carlos F. da Rocha; Fabio G. Cozman

doi:10.1007/3-540-36127-8_36

Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability

José Carlos F. da Rocha, Fabio G. Cozman

Resultado de la investigación: Capítulo del libro/informe/acta de congreso › Contribución a la conferencia

3 Citas (Scopus)

Resumen

© Springer-Verlag Berlin Heidelberg 2002. Probabilistic models and graph-based independence languages have often been combined in artificial intelligence research. The Bayesian network formalism is probably the best example of this type of association. In this article we focus on graphical structures that associate graphs with sets of probability measures — the result is referred to as a credal network. We describe credal networks and review an algorithm for evidential reasoning that we have recently developed. The algorithm substantially simplifies the computation of upper and lower probabilities by exploiting an independence assumption (strong independence) and a representation based on separately specified sets of probability measures. The algorithm is particularly efficient when applied to polytree structures. We then discuss a strategy for approximate reasoning in multi-connected networks, based on conditioning.

Idioma original	Inglés estadounidense
Título de la publicación alojada	Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability
Páginas	376-385
Número de páginas	10
ISBN (versión digital)	3540001247, 9783540001249
DOI	https://doi.org/10.1007/3-540-36127-8_36
Estado	Publicada - 1 ene 2002
Publicado de forma externa	Sí
Evento	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) - Duración: 1 ene 2018 → …

Serie de la publicación

Nombre	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volumen	2507
ISSN (versión impresa)	0302-9743

Conferencia

Conferencia	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Período	1/01/18 → …

Acceder al documento

10.1007/3-540-36127-8_36

Huella Profundice en los temas de investigación de 'Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability'. En conjunto forman una huella única.

Citar esto

da Rocha, J. C. F., & Cozman, F. G. (2002). Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability. En Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability (pp. 376-385). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2507). https://doi.org/10.1007/3-540-36127-8_36

da Rocha, José Carlos F. ; Cozman, Fabio G. / Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability. Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability. 2002. pp. 376-385 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ee02ddeefa6d45088129768506434de3,

title = "Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability",

abstract = "{\textcopyright} Springer-Verlag Berlin Heidelberg 2002. Probabilistic models and graph-based independence languages have often been combined in artificial intelligence research. The Bayesian network formalism is probably the best example of this type of association. In this article we focus on graphical structures that associate graphs with sets of probability measures — the result is referred to as a credal network. We describe credal networks and review an algorithm for evidential reasoning that we have recently developed. The algorithm substantially simplifies the computation of upper and lower probabilities by exploiting an independence assumption (strong independence) and a representation based on separately specified sets of probability measures. The algorithm is particularly efficient when applied to polytree structures. We then discuss a strategy for approximate reasoning in multi-connected networks, based on conditioning.",

author = "{da Rocha}, {Jos{\'e} Carlos F.} and Cozman, {Fabio G.}",

year = "2002",

month = jan,

day = "1",

doi = "10.1007/3-540-36127-8_36",

language = "American English",

isbn = "3540001247",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "376--385",

booktitle = "Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability",

note = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) ; Conference date: 01-01-2018",

}

da Rocha, JCF & Cozman, FG 2002, Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability. En Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2507, pp. 376-385, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 1/01/18. https://doi.org/10.1007/3-540-36127-8_36

Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability. / da Rocha, José Carlos F.; Cozman, Fabio G.

Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability. 2002. p. 376-385 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2507).

Resultado de la investigación: Capítulo del libro/informe/acta de congreso › Contribución a la conferencia

TY - GEN

T1 - Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability

AU - da Rocha, José Carlos F.

AU - Cozman, Fabio G.

PY - 2002/1/1

Y1 - 2002/1/1

N2 - © Springer-Verlag Berlin Heidelberg 2002. Probabilistic models and graph-based independence languages have often been combined in artificial intelligence research. The Bayesian network formalism is probably the best example of this type of association. In this article we focus on graphical structures that associate graphs with sets of probability measures — the result is referred to as a credal network. We describe credal networks and review an algorithm for evidential reasoning that we have recently developed. The algorithm substantially simplifies the computation of upper and lower probabilities by exploiting an independence assumption (strong independence) and a representation based on separately specified sets of probability measures. The algorithm is particularly efficient when applied to polytree structures. We then discuss a strategy for approximate reasoning in multi-connected networks, based on conditioning.

AB - © Springer-Verlag Berlin Heidelberg 2002. Probabilistic models and graph-based independence languages have often been combined in artificial intelligence research. The Bayesian network formalism is probably the best example of this type of association. In this article we focus on graphical structures that associate graphs with sets of probability measures — the result is referred to as a credal network. We describe credal networks and review an algorithm for evidential reasoning that we have recently developed. The algorithm substantially simplifies the computation of upper and lower probabilities by exploiting an independence assumption (strong independence) and a representation based on separately specified sets of probability measures. The algorithm is particularly efficient when applied to polytree structures. We then discuss a strategy for approximate reasoning in multi-connected networks, based on conditioning.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=70349778220&origin=inward

UR - https://www.scopus.com/inward/citedby.uri?partnerID=HzOxMe3b&scp=70349778220&origin=inward

U2 - 10.1007/3-540-36127-8_36

DO - 10.1007/3-540-36127-8_36

M3 - Conference contribution

SN - 3540001247

SN - 9783540001249

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 376

EP - 385

BT - Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability

T2 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Y2 - 1 January 2018

ER -

da Rocha JCF, Cozman FG. Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability. En Evidence propagation in credal networks: An exact algorithm based on separately specified sets of probability. 2002. p. 376-385. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/3-540-36127-8_36