Sorry, this article is only available in Русский.
 |
 |
UDC 681.3
PROSPECTIVE DEVELOPMENT OF COMPUTING BASED NONPOSITIONAL NEUROCOMPUTER
Gordenko Dmitryi Vladimirovich
Stavropol Agricultural University
Stavropol Agricultural University
Abstract
One of the most promising areas of computing, satisfying the requirements of the reform of education and science and technology are Neurocomputing, which are the basis of artificial neural networks.
For submission and processing of data can be used neurocomputers and nonpositional positional number system. Positional systems are conventional and aligning them with the neural networks used artificial methods that reduce the positive properties of neural networks associated with parallel computing. Nonpositional radix , particularly residual classes system is parallel to the system and provides a performance level parallelism of elementary operations , i.e. system of residual classes is the natural basis of data representation in neural networks , providing them with new features and capabilities. Combination of properties of neural networks and system of residual classes implement not only massive parallelism , but also allow us to develop new products resilient computing facilities . Parallel computing structures are an ideal basis for building fault tolerance neurocomputing means ultra high performance .
Keywords: artificial neural networks, neurocomputers, Neurocomputing, nonpositional positional number system
Category: Common rubric
Article reference:
Gordenko D.V. Prospective development of computing based nonpositional neurocomputer // Researches in Science. 2013. № 12 [Electronic journal]. URL: https://science.snauka.ru/en/2013/12/6473
View this article in Russian
© If you have found a violation of copyrights please notify us immediately by e-mail or feedback form.
Contact author (comments/reviews)
Write comment
You must authorise to write a comment.
Если Вы еще не зарегистрированы на сайте, то Вам необходимо зарегистрироваться: