Adaptation of the Regularization Parameters in the Nm-Delta Networks

• Autorzy: Gołąbek P. , Kosiński W.
• Języki publikacji: EN

Abstrakty

EN

The paper describes an application of regularization techniques to an automatic choice of parameters driving the learning process in the NM-Delta neural network architecture. The heterogeneous learning algorithm is identified as very similar to the Levenberg-Marquardt method but with a considerably smaller computational cost and different justification of parameter selection. The performance of the modified algorithm proves to be comparable with that of the Levenberg-Marquardt.

Słowa kluczowe

EN

NM-Delta; M-Delta; regularization; Levenberg-Marquardt; quasi-Newton

PL

regularyzacja; Levenberg-Marquardt

• Wydawca: Oficyna Wydawnicza Uniwersytetu Zielonogórskiego
• Czasopismo: International Journal of Applied Mathematics and Computer Science
• Rocznik: 2000
• Tom: Vol. 10, no 4
• Strony: 779--790
• Opis fizyczny: Bibliogr. 20 poz., wykr.

Twórcy

autor

Gołąbek P.

autor

Kosiński W.

• Radom Technical University Division of Electronics and Automatics ul. Malczewskiego 29, 26-600 Radom, Poland, Polish Academy of Sciences Division of Vision and Measurement Systems Institute of Fundamental Technological Research ul. Swietokrzysk,