Prior knowledge elicitation: The past, present, and future

Autores
Mikkola, Petrus; Martín, Osvaldo Antonio; Chandramoul, Suyog; Hartmann, Marcelo; Abril Pla, Oriol; Thomas, Owen; Pesonen, Henri; Corander, Jukka; Vehtari, Aki; Kaski, Samuel; Bürkner, Paul Christian; Klami, Arto
Año de publicación
2021
Idioma
inglés
Tipo de recurso
artículo
Estado
versión publicada
Descripción
Specification of the prior distribution for a Bayesian model is a central part of the Bayesian workflow for data analysis, but it is often difficult even for statistical experts. Prior elicitation transforms domain knowledge of various kinds into well-defined prior distributions, and offers a solution to the prior specification problem, in principle. In practice, however, we are still fairly far from having usable prior elicitation tools that could significantly influence the way we build probabilistic models in academia and industry. We lack elicitation methods that integrate well into the Bayesian workflow and perform elicitation efficiently in terms of costs of time and effort. We even lack a comprehensive theoretical framework for understanding different facets of the prior elicitation problem.Why are we not widely using prior elicitation? We analyze the state of the art by identifying a range of key aspects of prior knowledge elicitation, from properties of the modelling task and the nature of the priors to the form of interaction with the expert. The existing prior elicitation literature is reviewed and categorized in these terms. This allows recognizing under-studied directions in prior elicitation research, finally leading to a proposal of several new avenues to improve prior elicitation methodology.
Fil: Mikkola, Petrus. Aalto University; Finlandia
Fil: Martín, Osvaldo Antonio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - San Luis. Instituto de Matemática Aplicada de San Luis "Prof. Ezio Marchi". Universidad Nacional de San Luis. Facultad de Ciencias Físico, Matemáticas y Naturales. Instituto de Matemática Aplicada de San Luis "Prof. Ezio Marchi"; Argentina. Aalto University; Finlandia
Fil: Chandramoul, Suyog. Aalto University; Finlandia
Fil: Hartmann, Marcelo. University of Helsinki; Finlandia
Fil: Abril Pla, Oriol. University of Helsinki; Finlandia
Fil: Thomas, Owen. University of Oslo; Noruega
Fil: Pesonen, Henri. University of Oslo; Noruega
Fil: Corander, Jukka. University of Oslo; Noruega
Fil: Vehtari, Aki. Aalto University; Finlandia
Fil: Kaski, Samuel. Aalto University; Finlandia
Fil: Bürkner, Paul Christian. University Of Stuttgart; Alemania
Fil: Klami, Arto. University of Helsinki; Finlandia
Materia
prior elicitation
prior distribution
informative prior
Bayesian workflow
domain knowledge
Nivel de accesibilidad
acceso abierto
Condiciones de uso
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
Repositorio
CONICET Digital (CONICET)
Institución
Consejo Nacional de Investigaciones Científicas y Técnicas
OAI Identificador
oai:ri.conicet.gov.ar:11336/183197

id CONICETDig_41dd288b447976ba13f682559a10c922
oai_identifier_str oai:ri.conicet.gov.ar:11336/183197
network_acronym_str CONICETDig
repository_id_str 3498
network_name_str CONICET Digital (CONICET)
spelling Prior knowledge elicitation: The past, present, and futureMikkola, PetrusMartín, Osvaldo AntonioChandramoul, SuyogHartmann, MarceloAbril Pla, OriolThomas, OwenPesonen, HenriCorander, JukkaVehtari, AkiKaski, SamuelBürkner, Paul ChristianKlami, Artoprior elicitationprior distributioninformative priorBayesian workflowdomain knowledgehttps://purl.org/becyt/ford/1.1https://purl.org/becyt/ford/1Specification of the prior distribution for a Bayesian model is a central part of the Bayesian workflow for data analysis, but it is often difficult even for statistical experts. Prior elicitation transforms domain knowledge of various kinds into well-defined prior distributions, and offers a solution to the prior specification problem, in principle. In practice, however, we are still fairly far from having usable prior elicitation tools that could significantly influence the way we build probabilistic models in academia and industry. We lack elicitation methods that integrate well into the Bayesian workflow and perform elicitation efficiently in terms of costs of time and effort. We even lack a comprehensive theoretical framework for understanding different facets of the prior elicitation problem.Why are we not widely using prior elicitation? We analyze the state of the art by identifying a range of key aspects of prior knowledge elicitation, from properties of the modelling task and the nature of the priors to the form of interaction with the expert. The existing prior elicitation literature is reviewed and categorized in these terms. This allows recognizing under-studied directions in prior elicitation research, finally leading to a proposal of several new avenues to improve prior elicitation methodology.Fil: Mikkola, Petrus. Aalto University; FinlandiaFil: Martín, Osvaldo Antonio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - San Luis. Instituto de Matemática Aplicada de San Luis "Prof. Ezio Marchi". Universidad Nacional de San Luis. Facultad de Ciencias Físico, Matemáticas y Naturales. Instituto de Matemática Aplicada de San Luis "Prof. Ezio Marchi"; Argentina. Aalto University; FinlandiaFil: Chandramoul, Suyog. Aalto University; FinlandiaFil: Hartmann, Marcelo. University of Helsinki; FinlandiaFil: Abril Pla, Oriol. University of Helsinki; FinlandiaFil: Thomas, Owen. University of Oslo; NoruegaFil: Pesonen, Henri. University of Oslo; NoruegaFil: Corander, Jukka. University of Oslo; NoruegaFil: Vehtari, Aki. Aalto University; FinlandiaFil: Kaski, Samuel. Aalto University; FinlandiaFil: Bürkner, Paul Christian. University Of Stuttgart; AlemaniaFil: Klami, Arto. University of Helsinki; FinlandiaCornell University2021-12info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionhttp://purl.org/coar/resource_type/c_6501info:ar-repo/semantics/articuloapplication/pdfapplication/pdfhttp://hdl.handle.net/11336/183197Mikkola, Petrus; Martín, Osvaldo Antonio; Chandramoul, Suyog ; Hartmann, Marcelo ; Abril Pla, Oriol ; et al.; Prior knowledge elicitation: The past, present, and future; Cornell University; arXiv; 12-2021; 1-602331-8422CONICET DigitalCONICETenginfo:eu-repo/semantics/altIdentifier/url/https://arxiv.org/abs/2112.01380info:eu-repo/semantics/openAccesshttps://creativecommons.org/licenses/by-nc-sa/2.5/ar/reponame:CONICET Digital (CONICET)instname:Consejo Nacional de Investigaciones Científicas y Técnicas2025-09-03T10:05:04Zoai:ri.conicet.gov.ar:11336/183197instacron:CONICETInstitucionalhttp://ri.conicet.gov.ar/Organismo científico-tecnológicoNo correspondehttp://ri.conicet.gov.ar/oai/requestdasensio@conicet.gov.ar; lcarlino@conicet.gov.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:34982025-09-03 10:05:04.506CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicasfalse
dc.title.none.fl_str_mv Prior knowledge elicitation: The past, present, and future
title Prior knowledge elicitation: The past, present, and future
spellingShingle Prior knowledge elicitation: The past, present, and future
Mikkola, Petrus
prior elicitation
prior distribution
informative prior
Bayesian workflow
domain knowledge
title_short Prior knowledge elicitation: The past, present, and future
title_full Prior knowledge elicitation: The past, present, and future
title_fullStr Prior knowledge elicitation: The past, present, and future
title_full_unstemmed Prior knowledge elicitation: The past, present, and future
title_sort Prior knowledge elicitation: The past, present, and future
dc.creator.none.fl_str_mv Mikkola, Petrus
Martín, Osvaldo Antonio
Chandramoul, Suyog
Hartmann, Marcelo
Abril Pla, Oriol
Thomas, Owen
Pesonen, Henri
Corander, Jukka
Vehtari, Aki
Kaski, Samuel
Bürkner, Paul Christian
Klami, Arto
author Mikkola, Petrus
author_facet Mikkola, Petrus
Martín, Osvaldo Antonio
Chandramoul, Suyog
Hartmann, Marcelo
Abril Pla, Oriol
Thomas, Owen
Pesonen, Henri
Corander, Jukka
Vehtari, Aki
Kaski, Samuel
Bürkner, Paul Christian
Klami, Arto
author_role author
author2 Martín, Osvaldo Antonio
Chandramoul, Suyog
Hartmann, Marcelo
Abril Pla, Oriol
Thomas, Owen
Pesonen, Henri
Corander, Jukka
Vehtari, Aki
Kaski, Samuel
Bürkner, Paul Christian
Klami, Arto
author2_role author
author
author
author
author
author
author
author
author
author
author
dc.subject.none.fl_str_mv prior elicitation
prior distribution
informative prior
Bayesian workflow
domain knowledge
topic prior elicitation
prior distribution
informative prior
Bayesian workflow
domain knowledge
purl_subject.fl_str_mv https://purl.org/becyt/ford/1.1
https://purl.org/becyt/ford/1
dc.description.none.fl_txt_mv Specification of the prior distribution for a Bayesian model is a central part of the Bayesian workflow for data analysis, but it is often difficult even for statistical experts. Prior elicitation transforms domain knowledge of various kinds into well-defined prior distributions, and offers a solution to the prior specification problem, in principle. In practice, however, we are still fairly far from having usable prior elicitation tools that could significantly influence the way we build probabilistic models in academia and industry. We lack elicitation methods that integrate well into the Bayesian workflow and perform elicitation efficiently in terms of costs of time and effort. We even lack a comprehensive theoretical framework for understanding different facets of the prior elicitation problem.Why are we not widely using prior elicitation? We analyze the state of the art by identifying a range of key aspects of prior knowledge elicitation, from properties of the modelling task and the nature of the priors to the form of interaction with the expert. The existing prior elicitation literature is reviewed and categorized in these terms. This allows recognizing under-studied directions in prior elicitation research, finally leading to a proposal of several new avenues to improve prior elicitation methodology.
Fil: Mikkola, Petrus. Aalto University; Finlandia
Fil: Martín, Osvaldo Antonio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Conicet - San Luis. Instituto de Matemática Aplicada de San Luis "Prof. Ezio Marchi". Universidad Nacional de San Luis. Facultad de Ciencias Físico, Matemáticas y Naturales. Instituto de Matemática Aplicada de San Luis "Prof. Ezio Marchi"; Argentina. Aalto University; Finlandia
Fil: Chandramoul, Suyog. Aalto University; Finlandia
Fil: Hartmann, Marcelo. University of Helsinki; Finlandia
Fil: Abril Pla, Oriol. University of Helsinki; Finlandia
Fil: Thomas, Owen. University of Oslo; Noruega
Fil: Pesonen, Henri. University of Oslo; Noruega
Fil: Corander, Jukka. University of Oslo; Noruega
Fil: Vehtari, Aki. Aalto University; Finlandia
Fil: Kaski, Samuel. Aalto University; Finlandia
Fil: Bürkner, Paul Christian. University Of Stuttgart; Alemania
Fil: Klami, Arto. University of Helsinki; Finlandia
description Specification of the prior distribution for a Bayesian model is a central part of the Bayesian workflow for data analysis, but it is often difficult even for statistical experts. Prior elicitation transforms domain knowledge of various kinds into well-defined prior distributions, and offers a solution to the prior specification problem, in principle. In practice, however, we are still fairly far from having usable prior elicitation tools that could significantly influence the way we build probabilistic models in academia and industry. We lack elicitation methods that integrate well into the Bayesian workflow and perform elicitation efficiently in terms of costs of time and effort. We even lack a comprehensive theoretical framework for understanding different facets of the prior elicitation problem.Why are we not widely using prior elicitation? We analyze the state of the art by identifying a range of key aspects of prior knowledge elicitation, from properties of the modelling task and the nature of the priors to the form of interaction with the expert. The existing prior elicitation literature is reviewed and categorized in these terms. This allows recognizing under-studied directions in prior elicitation research, finally leading to a proposal of several new avenues to improve prior elicitation methodology.
publishDate 2021
dc.date.none.fl_str_mv 2021-12
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
http://purl.org/coar/resource_type/c_6501
info:ar-repo/semantics/articulo
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/11336/183197
Mikkola, Petrus; Martín, Osvaldo Antonio; Chandramoul, Suyog ; Hartmann, Marcelo ; Abril Pla, Oriol ; et al.; Prior knowledge elicitation: The past, present, and future; Cornell University; arXiv; 12-2021; 1-60
2331-8422
CONICET Digital
CONICET
url http://hdl.handle.net/11336/183197
identifier_str_mv Mikkola, Petrus; Martín, Osvaldo Antonio; Chandramoul, Suyog ; Hartmann, Marcelo ; Abril Pla, Oriol ; et al.; Prior knowledge elicitation: The past, present, and future; Cornell University; arXiv; 12-2021; 1-60
2331-8422
CONICET Digital
CONICET
dc.language.none.fl_str_mv eng
language eng
dc.relation.none.fl_str_mv info:eu-repo/semantics/altIdentifier/url/https://arxiv.org/abs/2112.01380
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
eu_rights_str_mv openAccess
rights_invalid_str_mv https://creativecommons.org/licenses/by-nc-sa/2.5/ar/
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv Cornell University
publisher.none.fl_str_mv Cornell University
dc.source.none.fl_str_mv reponame:CONICET Digital (CONICET)
instname:Consejo Nacional de Investigaciones Científicas y Técnicas
reponame_str CONICET Digital (CONICET)
collection CONICET Digital (CONICET)
instname_str Consejo Nacional de Investigaciones Científicas y Técnicas
repository.name.fl_str_mv CONICET Digital (CONICET) - Consejo Nacional de Investigaciones Científicas y Técnicas
repository.mail.fl_str_mv dasensio@conicet.gov.ar; lcarlino@conicet.gov.ar
_version_ 1842269890291957760
score 13.13397