Search
Some Methods of Adaptive Multilayer Neural Network Training
View/ Open document files
Date
2003Publisher
BSUIRCitation
Maniakov, N. Some Methods of Adaptive Multilayer Neural Network Training / Nikolaj Maniakov, Leonid Makhnist, Vladimir Rubanov // The 3nd International Conference on Neural Networks and Artificial Intelligence = Нейронные сети и искусственный интеллект : proceedings, Minsk, November 12–14, 2003 / Belarusian State University of Informatics and Radioelectronicsb ; ed.: Rauf Kh. Sadykhov [et al.]. – Minsk, 2003. – P. 109–115 : il. – Bibliogr.: p. 115 (2 titles).Abstract
Is proposed two new techniques for multilayer neural networks training. Its basic concept is based on the gradient descent method. For every methodic are showed formulas for calculation of the adaptive training steps. Matrix algorithmizations for all of these techniques are very helpful in its program realization.
Collection
Это произведение доступно по лицензии Creative Commons «Attribution-NonCommercial» («Атрибуция-Некоммерчески») 4.0 Всемирная.