Search

This paper introduces a novel synchronization scheme for fractional-order neural networks with time delays and reaction-diffusion terms via pinning control. We consider Caputo fractional derivatives, constant delays and distributed delays in our model. Based on the stability behavior, fractional inequalities and Lyapunov-type functions, several criteria are derived, which ensure the achievement of a synchronization for the drive-response systems. The obtained criteria are easy to test and are in the format of inequalities between the system parameters. Finally, numerical examples are presented to illustrate the results. The obtained criteria in this paper consider the effect of time delays as well as the reaction-diffusion terms, which generalize and improve some existing results. [ABSTRACT FROM AUTHOR]

Deep learning and Neural Network algorithms are a branch of Machine learning that can automatically identify patterns in the data, and then use the uncovered patterns to predict future data, or to perform other alternative kinds of decision making under unreliability. Neural Networks is a method of computing, which is based on a collection of large number of neural units, which acts as a biological brain and solve problems with large clusters of biological neurons connected by axons. Deep Learning algorithms are used to model high level abstractions in data. Digit Recognition is a combination of Deep Learning and Neural Network algorithms, which uses TensorFlow tool as an interface to develop a model. This paper describes the recognition of handwritten scanned digits by a system and displays the output as digital numbers by using Machine Learning methods with the help of TensorFlow tool. [ABSTRACT FROM AUTHOR]

This article examines the drive-response synchronization of a class of fractional order uncertain BAM (Bidirectional Associative Memory) competitive neural networks. By using the differential inclusions theory, and constructing a proper Lyapunov-Krasovskii functional, novel sufficient conditions are obtained to achieve global asymptotic stability of fractional order uncertain BAM competitive neural networks. This novel approach is based on the linear matrix inequality (LMI) technique and the derived conditions are easy to verify via the LMI toolbox. Moreover, numerical examples are presented to show the feasibility and effectiveness of the theoretical results. [ABSTRACT FROM AUTHOR]