Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
On the identifiability of mixtures-of-experts
Autore:
Jiang, W; Tanner, MA;
Indirizzi:
Northwestern Univ, Dept Stat, Evanston, IL 60208 USA Northwestern Univ Evanston IL USA 60208 Dept Stat, Evanston, IL 60208 USA
Titolo Testata:
NEURAL NETWORKS
fascicolo: 9, volume: 12, anno: 1999,
pagine: 1253 - 1258
SICI:
0893-6080(199911)12:9<1253:OTIOM>2.0.ZU;2-1
Fonte:
ISI
Lingua:
ENG
Soggetto:
EM ALGORITHM;
Keywords:
generalized linear models; indentifiability; invariant transformations; mixtures-of-experts;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
Citazioni:
15
Recensione:
Indirizzi per estratti:
Indirizzo: Jiang, W Northwestern Univ, Dept Stat, Evanston, IL 60208 USA NorthwesternUniv Evanston IL USA 60208 , Evanston, IL 60208 USA
Citazione:
W. Jiang e M.A. Tanner, "On the identifiability of mixtures-of-experts", NEURAL NETW, 12(9), 1999, pp. 1253-1258

Abstract

In mixtures-of-experts (ME) models, "experts" of generalized linear modelsare combined, according to a set of local weights called the "gating function". The invariant transformations of the ME probability density functionsinclude the permutations of the expert labels and the translations of the parameters in the gating functions. Under certain conditions, we show that the ME systems are identifiable if the experts are ordered and the gating parameters are initialized. The conditions are validated for Poisson, gamma,normal and binomial experts. (C) 1999 Elsevier Science Ltd. All rights reserved.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 19/01/20 alle ore 08:58:25