Naslov
How the Internal State Subnetwork Works

Autori
Jevtić, Dragan

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceedings of the IJCNN'99 / Boswell, Jonathan - Washington (MD) : Institute of Electrical and Electronics Engineers (IEEE), 1999, 3022-3024

Skup
International Joint Conference on Neural Networks, IJCNN'99

Mjesto i datum
Sjedinjene Američke Države, 10.07.2000. - 16.07.2000

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
rekurentna mreža; učenje u vremenu; parcijalne veze
(recurrrent network; learning in time; partial connections)

Sažetak
This paper presents a neural network architecture that performs classification using constructional simple and compact form of recurrent connections. The work was motivated by desire to decrease computational complexity and to maintain a greater degree of modularity in the neural network design. The main use of the network can be expected in various speech applications, in which the context information is crucial. A new method to extend the design of the Multy-layer Perceptron topology is introduced. The method uses an additional recurrent subnetwork module between two subsequent and processing layers of the feedforward network.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika

POVEZANOST RADA