University of Limerick Institutional Repository

Fully probabilistic design of hierarchical Bayesian models

DSpace Repository

Show simple item record

dc.contributor.author Quinn, Anthony
dc.contributor.author Kárný, MIroslav
dc.contributor.author Guy, Tatiana V
dc.date.accessioned 2017-01-04T15:21:02Z
dc.date.issued 2016
dc.identifier.uri http://hdl.handle.net/10344/5427
dc.description peer-reviewed en_US
dc.description.abstract The minimum cross-entropy principle is an established technique for design of an unknown distribution, processing linear functional constraints on the distribution. More generally, fully probabilistic design (FPD) chooses the distribution—within the knowledge-constrained set of possible distributions—for which the Kullback-Leibler divergence to the designer’s ideal distribution is minimized. These principles treat the unknown distribution deterministically. In this paper, fully probabilistic design is applied to hierarchical Bayesian models for the first time, yielding optimal design of a (possibly nonparametric) stochastic model for the unknown distribution. This equips minimum cross-entropy and FPD distributional estimates with measures of uncertainty. It enables robust choice of the optimal model, as well as randomization of this choice. The ability to process non-linear functional constraints in the constructed distribution significantly extends the applicability of these principles. Currently available FPD procedures for (a) merging of external knowledge, (b) approximate learning and stabilized forgetting, (c) decision strategy design, and (d) local adaptive control design, are unified for the first time via the hierarchical FPD framework of this paper. en_US
dc.language.iso eng en_US
dc.publisher Elsevier en_US
dc.relation.ispartofseries Information Sciences;369, pp. 532-547
dc.relation.uri http://dx.doi.org/10.1016/j.ins.2016.07.035
dc.rights This is the author’s version of a work that was accepted for publication in Information Sciences. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Information Sciences, 2016, 369, pp. 532-547, http://dx.doi.org/10.1016/j.ins.2016.07.035 en_US
dc.subject fully probabilistic design en_US
dc.subject ideal distribution en_US
dc.subject minimum cross-entropy principle en_US
dc.subject Bayesian conditioning en_US
dc.subject Kullback-Leibler divergence en_US
dc.subject Bayesian nonparametric modelling en_US
dc.title Fully probabilistic design of hierarchical Bayesian models en_US
dc.type info:eu-repo/semantics/article en_US
dc.type.supercollection all_ul_research en_US
dc.type.supercollection ul_published_reviewed en_US
dc.identifier.doi 10.1016/j.ins.2016.07.035
dc.contributor.sponsor SFI en_US
dc.relation.projectid 10/RFP/MTH2877 en_US
dc.relation.projectid 13-13502S en_US
dc.date.embargoEndDate 2018-07-14
dc.embargo.terms 2018-07-14 en_US
dc.rights.accessrights info:eu-repo/semantics/openAccess en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search ULIR


Browse

My Account

Statistics