Statistical relational learning: Difference between revisions

Revision as of 14:59, 29 April 2020

Revision as of 23:48, 10 July 2020

Line 1:

'''Statistical relational learning''' ('''SRL''') is a subdiscipline of [[artificial intelligence]] and [[machine learning]] that is concerned with [[___domain model]]s that exhibit both [[uncertainty]] (which can be dealt with using statistical methods) and complex, [[relation (mathematics)|relational]] structure.
<ref name=Getoor2007>[[Lise Getoor]] and [[Ben Taskar]]getoor:book07 ''[https://books.google.com/books?id=lSkIewOw2WoC&printsec=frontcover#v=onepage&q&f=false Introduction to statistical relational learning]'', MIT Press, 2007</ref>
<ref>Ryan A. Rossi, Luke K. McDowell, David W. Aha, and Jennifer Neville, "[httpname=rossi://www.jair.org/media/3659/live-3659-6589-jair.pdfjair12 Transforming Graph Data for Statistical Relational Learning.]" ''Journal of Artificial Intelligence Research (JAIR)'', '''Volume 45''' (2012), pp. 363-441.</ref>
Note that SRL is sometimes called Relational Machine Learning (RML) in the literature. Typically, the [[knowledge representation]] formalisms developed in SRL use (a subset of) [[first-order logic]] to describe relational properties of a ___domain in a general manner ([[universal quantification]]) and draw upon [[probabilistic graphical model]]s (such as [[Bayesian network]]s or [[Markov network]]s) to model the uncertainty; some also build upon the methods of [[inductive logic programming]]. Significant contributions to the field have been made since the late 1990s.
<ref>{{citation_needed|date name=Decembergetoor:book07 2016}}</ref>

As is evident from the characterization above, the field is not strictly limited to learning aspects; it is equally concerned with [[Semantic_reasoner|reasoning]] (specifically [[statistical inference|probabilistic inference]]) and [[knowledge representation]]. Therefore, alternative terms that reflect the main foci of the field include ''statistical relational learning and reasoning'' (emphasizing the importance of reasoning) and ''first-order probabilistic languages'' (emphasizing the key properties of the languages with which models are represented).

== Canonical tasks ==

A number of canonical tasks are associated with statistical relational learning, the most common ones being <ref>Matthew Richardson and Pedro Domingos, [httpname=richardson://www.cs.washington.edu/homes/pedrod/papers/mlj05.pdfml06 "Markov Logic Networks.]" ''Machine Learning'', '''62''' (2006), pp. 107–136.</ref>.

* [[collective classification]], i.e. the (simultaneous) [[classification (machine learning)|prediction of the class]] of several objects given objects' attributes and their relations

Line 15 ⟶ 19:

{{More footnotes|date=June 2011}}

One of the fundamental design goals of the representation formalisms developed in SRL is to abstract away from concrete entities and to represent instead general principles that are intended to be universally applicable. Since there are countless ways in which such principles can be represented, many representation formalisms have been proposed in recent years.<ref name=Getoor2007getoor:book07 /> In the following, some of the more common ones are listed in alphabetical order:

* [[Bayesian logic program]]

Line 22 ⟶ 26:

* [[Markov logic network]]s

* [[Multi-entity Bayesian network]]

* Probabilistic relational model – a Probabilistic Relational Model (PRM) is the counterpart of a [[Bayesian network]] in statistical relational learning.<ref>Friedman N, Getoor L, Koller D, Pfeffer A. (1999) [https://www.biostat.wisc.edu/~page/lprm-ijcai99.pdf "Learning probabilistic relational models"]. In: ''International joint conferences on artificial intelligence'', 1300–09</ref><ref>Teodor Sommestad, Mathias Ekstedt, Pontus Johnson (2010) "A probabilistic relational model for security risk analysis", ''Computers & Security'', 29 (6), 659-679 {{DOI|10.1016/j.cose.2010.02.002}}</ref>

<ref name=friedman:ijcai99 />

<ref name=sommestad:compsec10 />

* [[Probabilistic soft logic]]

* [[Recursive random field]]

Line 44 ⟶ 50:

== References ==

{{reflist}}|

refs=

<ref name=getoor:book07>

{{cite book |last1=Getoor |first1=Lise |last2=Taskar |first2=Ben |author-link1=Lise Getoor |author-link2=Ben Taskar |date=2007 |title=Introduction to Statistical Relational Learning |url=https://linqs.github.io/linqs-website/publications/#id:getoor-book07 |publisher=MIT Press |isbn=978-0262072885}}

</ref>

<ref name=rossi:jair12>

Ryan A. Rossi, Luke K. McDowell, David W. Aha, and Jennifer Neville, "[http://www.jair.org/media/3659/live-3659-6589-jair.pdf Transforming Graph Data for Statistical Relational Learning.]" ''Journal of Artificial Intelligence Research (JAIR)'', '''Volume 45''' (2012), pp. 363-441.

</ref>

<ref name=richardson:ml06>

Matthew Richardson and Pedro Domingos, [http://www.cs.washington.edu/homes/pedrod/papers/mlj05.pdf "Markov Logic Networks.]" ''Machine Learning'', '''62''' (2006), pp. 107–136.

</ref>

<ref name=friedman:ijcai99>

Friedman N, Getoor L, Koller D, Pfeffer A. (1999) [https://www.biostat.wisc.edu/~page/lprm-ijcai99.pdf "Learning probabilistic relational models"]. In: ''International joint conferences on artificial intelligence'', 1300–09

</ref>

<ref name=sommestad:compsec10>

Teodor Sommestad, Mathias Ekstedt, Pontus Johnson (2010) "A probabilistic relational model for security risk analysis", ''Computers & Security'', 29 (6), 659-679 {{DOI|10.1016/j.cose.2010.02.002}}

</ref>

}}