Журнал «Современная Наука»

Russian (CIS)English (United Kingdom)
 MOSCOW +7(495)-142-86-81

Application of echo state neural networks for time series forecasting

Shelonik A.   (Bauman Moscow State Technical University)

Smirnov A.   (Bauman Moscow State Technical University)

Kalistratov A.   (Bauman Moscow State Technical University )

Koldobskiy V.   (Bauman Moscow State Technical University)

The architecture of an echo state neural network is described. An echo state neural network is a recurrent neural network. An echo state network allows effective time series forecasting. Having low computational requirements, an echo state network shows results that are highly dependent on randomly generated initial values. A solution of this dependency problem is proposed. Usage of the multilayer architecture of the network and a genetic algorithm for acquiring network parameters is described. Echo state neural network’s performance is compared with other networks’ results.

Keywords:neural network, neural network architecture, time series forecasting
 

Read the full article …

Citation link:
Shelonik A. , Smirnov A. , Kalistratov A. , Koldobskiy V. Application of echo state neural networks for time series forecasting // Современная наука: актуальные проблемы теории и практики. Серия: Естественные и Технические Науки. -2017. -№05. -С. 70-74
LEGAL INFORMATION:
Reproduction of materials is permitted only for non-commercial purposes with reference to the original publication. Protected by the laws of the Russian Federation. Any violations of the law are prosecuted.
© ООО "Научные технологии"

Terms Publications

Issues of the journals (Archive)

ECONOMICS AND LAW
HUMANITIES
NATURAL AND TECHNICAL SCIENCES
COGNITION

Серия - Natural and Technical Sciences