In this article, we establish the superstability of diff erential equations of second order with boundary conditions or with initial conditions as well as the superstability of di fferential equations of higher order with initial... more
In this article, we establish the superstability of diff erential equations of second order with boundary conditions or with initial conditions as well as the superstability of di fferential equations of higher order with initial conditions.
Artificial Neural Networks (ANN's) are largely used in applications involving classification or functions approximation. It has been proved that several classes of ANN such as Multilayer Radial-Basis Function Networks (RBFN) are... more
Artificial Neural Networks (ANN's) are largely used in applications involving classification or functions approximation. It has been proved that several classes of ANN such as Multilayer Radial-Basis Function Networks (RBFN) are universal function approximators . Therefore, they are widely used for function approximation . In this paper ,we examine the similarities and differences between RBFNNs compare the performance of learning with each representation applied to the interpolation problem. Nonetheless, this paper should help the reader to understand which basis function and which efficient method should be employed for particular reconstruction problem. It should also encourage the reader to consult the literature pointed out in the bibliography for further studying.
In this paper ,we examine the similarities and differences between RBFNNs and compare the performance of learning, then we applied to the interpolation problem by using data of blood pressure disease which taken from health office in... more
In this paper ,we examine the similarities and differences between RBFNNs and compare the performance of learning, then we applied to the interpolation problem by using data of blood pressure disease which taken from health office in diwaniya city .
In this paper, we show the degree of approximation by a single hidden layer feed forward model with n units in the hidden layer is bounded below by the degree of approximation by a linear combination of n ridge functions. We prove that... more
In this paper, we show the degree of approximation by a single hidden layer feed forward model with n units in the hidden layer is bounded below by the degree of approximation by a linear combination of n ridge functions. We prove that there exists an analytic, strictly monotone, sigmoidal activation function for which this lower bound is essentially attained. Also we extend the Kolmogorov’s existence theorem to be apply at any compact set, (i.e., closed and bounded set) also we prove that a FFNN with one hidden layer can uniformly approximate any continuous function of several variable, f(x1, x2, …, xn), which is defined in compact set to any required accuracy.
We establish the generalized superstability of differential equations of nth-order with initial conditions and investigate the generalized superstability of differential equations of second order in the form of y′′(x) + p(x)y′(x)+q(x)y(x)... more
We establish the generalized superstability of differential equations of nth-order with initial conditions and investigate the generalized superstability of differential equations of second order in the form of y′′(x) + p(x)y′(x)+q(x)y(x) = 0 and the superstability of linear differential equations with constant coefficients with initial conditions.
In this article, we establish the superstability of differential equations of second order with boundary conditions or with initial conditions as well as the superstability of differential equations of higher order with initial conditions.
