Pregled bibliografske jedinice broj: 427157
Discrete Time Neural Network Synthesis Using Input Activation Functions
Discrete Time Neural Network Synthesis Using Input Activation Functions // Proc. of 1994 IEEE International Conference on Systems, Man, and Cybernetics, Vol. 3
San Antonio (TX): Institute of Electrical and Electronics Engineers (IEEE), 1994. str. 2516-2521 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 427157 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Discrete Time Neural Network Synthesis Using Input Activation Functions
Autori
Novaković, Branko
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proc. of 1994 IEEE International Conference on Systems, Man, and Cybernetics, Vol. 3
/ - San Antonio (TX) : Institute of Electrical and Electronics Engineers (IEEE), 1994, 2516-2521
Skup
1994 IEEE International Conference on Systems, Man, and Cybernetics
Mjesto i datum
San Antonio (TX), Sjedinjene Američke Države, 02.10.1994. - 05.10.1994
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
Discrete Time Systems; Neural Networks Synthesis; Input Activation Functions
Sažetak
In this paper the combinations of input and output activation functions for Neural Networks (NN) are employed. In that sense, it is presented the synthesis procedure for determination of input weight matrices. The obtained NN structure is employed for construction of universal logical units. Also, the same structure of NN has been used for synthesis of NN for robot control.
Izvorni jezik
Engleski
Znanstvena područja
Elektrotehnika, Strojarstvo
POVEZANOST RADA
Ustanove:
Fakultet strojarstva i brodogradnje, Zagreb
Profili:
Branko Novaković
(autor)