Convergence Properties of Beale - Powell
نویسندگان
چکیده
The Beale-Powell restart algorithm is highly useful for large-scale unconstrained optimization. In this paper, we use an example to show that the algorithm may fail to converge. The global convergence of a slightly modiied algorithm is proved.
منابع مشابه
کنترل بهینة شار حرارتی سطحی در یک جسم دوبعدی با خواص وابسته به دما
In this paper the optimal control of boundary heat flux in a 2-D solid body with an arbitrary shape is performed in order to achieve the desired temperature distribution at a given time interval. The boundary of the body is subdivided into a number of components. On each component a time-dependent heat flux is applied which is independent of the others. Since the thermophysical properties are t...
متن کاملAn Efficient Numerical Method to Solve the Boundary Layer Flow of an Eyring-Powell Non-Newtonian Fluid
In this paper, the boundary layer flow of an Eyring-Powell non-Newtonian fluid over a linearly stretching sheet is solved using the combination of the quasilinearization method and the Fractional order of Rational Chebyshev function (FRC) collocation method on a semi-infinite domain. The quasilinearization method converts the equation into a sequence of linear equations then, using the FRC coll...
متن کاملConjugate Gradient Methods in Training Neural Networks
Training of artificial neural networks is normally a time consuming task due to iterative search imposed by the implicit nonlinearity of the network behavior. To tackle the supervised learning of multilayer feed forward neural networks, the backpropagation algorithm has been proven to be one of the most successful neural network algorithm. Although backpropagation training has proved to be effi...
متن کاملQuadratic Convergence of Vortex Methods
We prove quadratic convergence for two-dimensional vortex methods with positive cutoffs. The result is established for flows with initial vorticity three times continuously differentiable and compact support. The proof is based on a refined version of a convergence result. Introduction. The purpose of this paper is to prove that vortex methods with positive cutoffs can converge quadratically if...
متن کاملAdvanced Neural Network Learning Applied to Pulping Modeling
This paper reports work done to improve the modeling of complex processes when only small experimental data sets are available. Neural networks are used to capture the nonlinear underlying phenomena contained in the data set and to partly eliminate the burden of having to specify completely the structure of the model. Two different types of neural networks were used for the application of pulpi...
متن کامل