В последние десятилетия в мире бурн

В последние десятилетия в мире бурно развивается новая прикладная область математики, специализирующаяся на искусственных нейронных сетях. Актуальность исследований в этом направлении подтверждается массой различных применений нейросетей. Это автоматизация процессов распознавания образов, адаптивное управление, аппроксимация функционалов, прогнозирование, создание экспертных систем, организация ассоциативной памяти и многие другие приложения. С помощью нейросетей можно, например, предсказывать показатели биржевого рынка, выполнять распознавание оптических или звуковых сигналов, создавать самообучающиеся системы, способные управлять автомашиной при парковке или синтезировать речь по тексту. В то время как на западе применение НС уже достаточно обширно, у нас это еще в некоторой степени экзотика – российские фирмы, использующие НС в практических целях, наперечет.
Широкий круг задач, решаемый нейронными сетями, не позволяет в настоящее время создавать универсальные, мощные сети, вынуждая разрабатывать специализированные сети, функционирующие по различным алгоритмам. Тем не менее, тенденции развития нейросетей растут с каждым годом.
Цель моей работы – разбор базовых понятий, связанных с изучением нейронных сетей, а также выявление перспектив развития.

Результаты (английский) 1:

In recent decades, the world is booming new applied mathematics, specializing in artificial neural networks. The relevance of the research in this direction is confirmed by weighing various applications of neural networks. This automation of the processes of pattern recognition, adaptive control, approximation of functionals, forecasting, creating expert systems, associative memory, and many other applications. With the help of neural networks can be, for example, to predict the performance of the stock market, perform OCR optical or audible signals create self-trained systems capable of control the car when parking or synthesize it in the text. While in the West the use of NS is already quite extensively, we are still somewhat exotic-Russian firms using NS for practical purposes, politicians.A wide range of tasks, kept neural networks do not allow now to create versatile, powerful network, forcing him to develop specialized network operating on various algorithms. Nevertheless, the trends in the development of neural networks are growing every year.The purpose of my work is to parse basic concepts related to the study of neural networks, as well as to identify prospects.

Результаты (английский) 2:

In recent decades, the world's booming new area of applied mathematics, specializing in artificial neural networks. The urgency of research in this area is confirmed by a mass of different neural networks applications. This automated pattern recognition processes, adaptive control, functional approximation, forecasting, development of expert systems, the organization of associative memory, and many other applications. With the help of neural networks can, for example, to predict the performance of the exchange market, to carry out detection of optical or acoustic signals, create a self-learning system that can operate a motor vehicle when parking or synthesize it in the text. While in the West NA application already quite extensive, we is still somewhat exotic - Russian firms that use HC for practical purposes, all without exception.
A wide range of tasks solved by neural networks do not allow at the moment to create a versatile, powerful network, forcing developing specialized networks that operate on different algorithms. However, trends in the development of neural networks are growing every year.
The purpose of my work - the analysis of the basic concepts related to the study of neural networks, as well as identifying development prospects.

Результаты (английский) 3:

in the last decade in the world and develops new application area of mathematics, specializing in artificial neural networks. the research in this direction is a variety of applications нейросетей. this is the automation of processes, pattern recognition, adaptive control, approximation функционалов, forecasting, establishment of expert systems, the associative memory and many other applications. with the help of нейросетей can, for example, to predict the performance of stock market to carry out recognition of optical or acoustic signals, create самообучающиеся systems that can control the vehicle in the parking lot or synthesized speech to text. while in the west the use of ns is enough also, it's still somewhat exotic, russian firms using ns in practical purposes, certainly.a wide range of tasks, решаемый нейронными networks, at present does not allow to create a universal, powerful network, forcing to develop specialized networks operating on the various algorithms. however, the trends in the development of нейросетей grow every year.the aim of my work is the analysis of basic concepts related to the study of neural networks, and the prospects for development.