-
Asymptotic Behaviour of Gradient Learning Algorithms in Neural Network Models for the Identification of Nonlinear Systems
Valerii N. Azarskov,
Dmytro P. Kucherov,
Sergii A. Nikolaienko,
Leonid S. Zhiteckii
Issue:
Volume 1, Issue 1, August 2015
Pages:
1-10
Received:
14 June 2015
Accepted:
28 July 2015
Published:
29 July 2015
Abstract: This paper deals with studying the asymptotical properties of multilayer neural networks models used for the adaptive identification of wide class of nonlinearly parameterized systems in stochastic environment. To adjust the neural network’s weights, the standard online gradient type learning algorithms are employed. The learning set is assumed to be infinite but bounded. The Lyapunov-like tool is utilized to analyze the ultimate behaviour of learning processes in the presence of stochastic input variables. New sufficient conditions guaranteeing the global convergence of these algorithms in the stochastic frameworks are derived. The main their feature is that they need no a penalty term to achieve the boundedness of weight sequence. To demonstrate asymptotic behaviour of the learning algorithms and support the theoretical studies, some simulation examples are also given
Abstract: This paper deals with studying the asymptotical properties of multilayer neural networks models used for the adaptive identification of wide class of nonlinearly parameterized systems in stochastic environment. To adjust the neural network’s weights, the standard online gradient type learning algorithms are employed. The learning set is assumed to ...
Show More
-
Study of the Nature and Structure of Corporate Culture with Neural Network Models
Sergiy V. Kovalevskyy,
Ludmila V. Kosheva
Issue:
Volume 1, Issue 1, August 2015
Pages:
11-22
Received:
14 June 2015
Accepted:
28 July 2015
Published:
29 July 2015
Abstract: The article contains the new materials reflecting application of neural network models in designing of innovative processes in pedagogic. Such approach, according to authors' point of view, is actual because it is very important to provide quantitative estimations along with quality standards for management of pedagogical processes. They allow to reveal tendencies of innovative pedagogical approaches in education of new generation of young men, but also to correspond to their aspirations, supporting the positive of their socioeconomic influence in every possible way
Abstract: The article contains the new materials reflecting application of neural network models in designing of innovative processes in pedagogic. Such approach, according to authors' point of view, is actual because it is very important to provide quantitative estimations along with quality standards for management of pedagogical processes. They allow to r...
Show More
-
Energy-saving Technologies and Research I Using Neural Network Models
Sergiy V. Kovalevskyy,
Ekaterina A. Zavgorodnyaya
Issue:
Volume 1, Issue 1, August 2015
Pages:
23-28
Received:
14 June 2015
Accepted:
28 July 2015
Published:
29 July 2015
Abstract: In article he results of researches in different aspects of surface treatment are submitted. New type of the process of stabilization of residual pressure is considered. Also new ways of hardening of details are investigated and compared. The problem of new tools and the question of its more effective using are considered. The way of neural network modeling for data processing was used for in all experiments
Abstract: In article he results of researches in different aspects of surface treatment are submitted. New type of the process of stabilization of residual pressure is considered. Also new ways of hardening of details are investigated and compared. The problem of new tools and the question of its more effective using are considered. The way of neural network...
Show More
-
Improving of Procedures for Preparing of Training Set for Neural Networks
Veniamin B. Gitis,
Tatyana P. Gitis
Issue:
Volume 1, Issue 1, August 2015
Pages:
29-32
Received:
14 June 2015
Accepted:
28 July 2015
Published:
1 August 2015
Abstract: In the article procedure of rough-down of information is examined for teaching of neuron networks. Shown, that exists problem of normalization of ordinals of variables in part of their internal levels. The improved chart of normalization, allowing setting ponder ability both ordinal of variable on the whole and its separate levels, is offered to application. Reverse normalization formulas over are also brought for interpretation of gravimetric coefficients of neurons
Abstract: In the article procedure of rough-down of information is examined for teaching of neuron networks. Shown, that exists problem of normalization of ordinals of variables in part of their internal levels. The improved chart of normalization, allowing setting ponder ability both ordinal of variable on the whole and its separate levels, is offered to ap...
Show More
