Skip Navigation Links
Strona główna/Home
Rada Naukowa/Scientific Council
Redakcja/Editorial office
Dla Autorów/For Authors
Recenzenci/Reviews
Wyszukaj/Search
Archiwum/Archives
Kontakt/Contact
Prenumerata/Subscription

Jesteś gościem nr:
533710
   

Szczegóły artykułu:

Wydawnictwo: Academic Journals Poznan University of Technology

Numer: 84/2015 Str: 113


Autorzy: Stanisław Płaczek


Tytuł: Decomposition and the principle of interaction prediction in hierarchical structure of learning algorithm of ANN


Streszczenie: For the most popular ANN structure with one hidden layer, decomposition is done into two sub-networks. These sub-networks form the first level of the hierarchical structure. On the second level, the coordinator is working with its own target function. In the hierarchical systems theory three coordination strategies are defined. For the ANN learning algorithm the most appropriate is the coordination by the principle of interaction prediction. Implementing an off-line algorithm in all sub-networks makes the process of weight coefficient modification more stable. In the article, the quality and quantity characteristics of a coordination algorithm and the result of the learning algorithm for all sub-networks are shown. Consequently, the primary ANN achieves the global minimum during the learning process.


Słowa kluczowe: Artificial Neural Network, hierarchy, decomposition, coordination, coordination principle.


[PDF]