In this paper, we derive and prove the stability bounds of the momentum coefficient µ and the learning rate ? of the back propagation updating rule in Artificial Neural Networks .The theoretical upper bound of learning rate ? is derived and its practical approximation is obtained

Nowadays, information systems constitute a crucial part of organizations; by losing security, these organizations will lose plenty of competitive advantages as well. The core point of information security (InfoSecu) is risk management. There are a great deal of research works and standards in security risk management (ISRM) including NIST 800-30 and ISO/IEC 27005. However, only few works of research focus on InfoSecu risk reduction, while the standards explain general principles and guidelines. They do not provide any implementation details regarding ISRM; as such reducing the InfoSecu risks in uncertain environments is painstaking. Thus, this paper applied a genetic algorithm (GA) for InfoSecu risk reduction in uncertainty. Finally, the ef

Features is the description of the image contents which could be corner, blob or edge. Corners are one of the most important feature to describe image, therefore there are many algorithms to detect corners such as Harris, FAST, SUSAN, etc. Harris is a method for corner detection and it is an efficient and accurate feature detection method. Harris corner detection is rotation invariant but it isn’t scale invariant. This paper presents an efficient harris corner detector invariant to scale, this improvement done by using gaussian function with different scales. The experimental results illustrate that it is very useful to use Gaussian linear equation to deal with harris weakness.

The concern of this article is the calculation of an upper bound of second Hankel determinant for the subclasses of functions defined by Al-Oboudi differential operator in the unit disc. To study special cases of the results of this article, we give particular values to the parameters A, B and λ

The proposed design of neural network in this article is based on new accurate approach for training by unconstrained optimization, especially update quasi-Newton methods are perhaps the most popular general-purpose algorithms. A limited memory BFGS algorithm is presented for solving large-scale symmetric nonlinear equations, where a line search technique without derivative information is used. On each iteration, the updated approximations of Hessian matrix satisfy the quasi-Newton form, which traditionally served as the basis for quasi-Newton methods. On the basis of the quadratic model used in this article, we add a new update of quasi-Newton form. One innovative features of this form's is its ability to estimate the energy functio