Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 3 of 3
Full-Text Articles in Entire DC Network
Function Approximation Using A Sinc Neural Network, Wael R. Elwasif, Laurene V. Fausett
Function Approximation Using A Sinc Neural Network, Wael R. Elwasif, Laurene V. Fausett
Mathematics and System Engineering Faculty Publications
Neural networks for function approximation are the basis of many applications. Such networks often use a sigmoidal activation function (e.g. tanh) or a radial basis function (e.g. gaussian). Networks have also been developed using wavelets. In this paper, we present a neural network approximation of functions of a single variable, using sinc functions for the activation functions on the hidden units. Performance of the sinc network is compared with that of a tanh network with the same number of hidden units. The sinc network generally learns the desired input-output mapping in significantly fewer epochs, and achieves a much lower total …
On Applications Of Excess Level Processes To (N,D)-Policy Bulk Queueing Systems, Jewgeni H. Dshalalow
On Applications Of Excess Level Processes To (N,D)-Policy Bulk Queueing Systems, Jewgeni H. Dshalalow
Mathematics and System Engineering Faculty Publications
The paper deals with queueing systems in which N- and D-policies are combined into one. This means that an idle or vacationing server will resume his service if the queueing or workload process crosses some specified fixed level N or D, respectively. For the proposed (N,D)-policy we study the queueing processes in models with and without server vacations, with compound Poisson input, and with generally distributed service and vacation periods. The analysis of the models is essentially based on fluctuation techniques for two-dimensional marked counting processes newly developed by the author. The results enable us to arrive at stationary distributions …
On The Structure Of The Deflagration For The Generalized Reaction-Rate Model, William B. Bush, L. Krishnamurthy
On The Structure Of The Deflagration For The Generalized Reaction-Rate Model, William B. Bush, L. Krishnamurthy
Mathematics and System Engineering Faculty Publications
The structure of the deflagration is examined by means of an asymptotic analysis of the physical-plane boundary-value problem, with Lewis-Semenov number unity, in the limit of the activation-temperature ratio, β = Ta/Tb, greater than order unity, for the generalized reaction-rate-model case of (1) the heat- addition-temperature ratio, α = (Tb — Tu)/Tu, of order unity [where Ta, Tb, and Tu are the activation, adiabatic-flame (and/or burned-gas), and unburned-gas temperatures, respectively]; and (2) the exponent, a, which characterizes the pre-exponential thermal dependence of the reaction-rate term, unity. This examination indicates that the deflagration has a four-region structure. To obtain a uniformly …