Catalogo Articoli (Spogli Riviste)
OPAC HELP
Titolo: Selforganizing mixture networks for probability density estimation
Autore: Yin, HJ; Allinson, NM;
 Indirizzi:
 Univ Manchester, Inst Sci & Technol, Dept Elect Engn & Elect, Manchester M60 1QD, Lancs, England Univ Manchester Manchester Lancs England M60 1QD M60 1QD, Lancs, England
 Titolo Testata:
 IEEE TRANSACTIONS ON NEURAL NETWORKS
fascicolo: 2,
volume: 12,
anno: 2001,
pagine: 405  411
 SICI:
 10459227(200103)12:2<405:SMNFPD>2.0.ZU;26
 Fonte:
 ISI
 Lingua:
 ENG
 Soggetto:
 EM ALGORITHM; LIKELIHOOD;
 Keywords:
 density estimation; expectationmaximization; (EM) algorithm; maximum likelihood; mixture distribution; selforganizing maps; unsupervised learning;
 Tipo documento:
 Article
 Natura:
 Periodico
 Settore Disciplinare:
 Engineering, Computing & Technology
 Citazioni:
 23
 Recensione:
 Indirizzi per estratti:
 Indirizzo: Yin, HJ Univ Manchester, Inst Sci & Technol, Dept Elect Engn & Elect, Manchester M60 1QD, Lancs, England Univ Manchester Manchester Lancs England M60 1QD , Lancs, England



 Citazione:
 H.J. Yin e N.M. Allinson, "Selforganizing mixture networks for probability density estimation", IEEE NEURAL, 12(2), 2001, pp. 405411
Abstract
A selforganizing mixture network (SOMN) is derived for learning arbitrarydensity functions, The network minimizes the KullbackLeibler information metric by means of stochastic approximation methods. The density functions are modeled as mixtures of parametric distributions A mixture needs not to be homogenous, i.e., it can have different density profiles. The first layer of the network is similar to Kohonen's selforganizing map (SOM), but with the parameters of the component densities as the learning weights, The winning mechanism is based on maximum posterior probability, and updating of the weights is limited to a small neighborhood around the winner. The second layer accumulates the responses of these local nodes, weighted by the learned mixing parameters, The network possesses a simple structure and computational form, yet yields fast and robust convergence. The network has a generalization ability due to the relative entropy criterion used, Applications to density profile estimation and pattern classification are presented. The SOMN can also provide an insight to the role of neighborhood function used in the SOM.
ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 06/04/20 alle ore 08:26:38