Statistical relational learning: Difference between revisions

Revision as of 23:08, 26 January 2017

Revision as of 23:12, 26 January 2017

Line 1:

'''Statistical relational learning''' ('''SRL''') is a subdiscipline of [[artificial intelligence]] and [[machine learning]] that is concerned with [[Domain model|___domain models]] 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: Introduction to statistical relational learning, MIT Press, 2007</ref><ref>Ryan A. Rossi, Luke K. McDowell, David W. Aha, and Jennifer Neville, [dx.doi.org/10.1613/jair.3659 "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=December 2016}}</ref>

As is evident from the characterization above, the field is not strictly limited to learning aspects; it is equally concerned with [[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).

Line 15:

{{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>Lise Getoorname=Getoor2007 and Ben Taskar: Introduction to statistical relational learning, MIT Press, 2007</ref> In the following, some of the more common ones are listed in alphabetical order:

* [[Bayesian logic program]]