Stability and convergence of general multistep and multivalue methods with variable step size.

Thesis--University of Illinois.

A study of the non-uniform convergence of the inverse of a doubly-infinite matrix associated with a boundary value problem in a waveguide /

"Contract AF33(616)-6079 Project No. 9-(13-6278), Task No. 40572. Sponsored by: Aeronautical Systems Division"

Stability and convergence of variable order multistep methods,

Includes bibliographical references.

Conservation laws and variational principles in general relativity /

"April 23, 1962."

Variational and Lagrangian thermodynamics of thermal convection-- fundamental shortcomings of the heat-transfer coefficient.

Mode of access: Internet.

On the convergence of reinforcement procedures in simple perceptors (Project Para)

Mode of access: Internet.

On the variational principles of plasticity.

At head of title: Office of Naval Research, Contract N7onr-358 T.O.1 NR-041-032.

Variational analysis of ablation for variable properties

"Work was performed as part of a Cornell Aeronautical Laboratory Internal Research Project."

One Hundred Years of the Concert Marimba: American and Japanese Innovation and Convergence, 1915 to 2014

Thesis (D.M.A.)--University of Washington, 2016-06

Using the EM Algorithm to Train Neural Networks: Misconceptions and a New Algorithm for Multiclass Classification

The expectation-maximization (EM) algorithm has been of considerable interest in recent years as the basis for various algorithms in application areas of neural networks such as pattern recognition. However, there exists some misconceptions concerning its application to neural networks. In this paper, we clarify these misconceptions and consider how the EM algorithm can be adopted to train multilayer perceptron (MLP) and mixture of experts (ME) networks in applications to multiclass classification. We identify some situations where the application of the EM algorithm to train MLP networks may be of limited value and discuss some ways of handling the difficulties. For ME networks, it is reported in the literature that networks trained by the EM algorithm using iteratively reweighted least squares (IRLS) algorithm in the inner loop of the M-step, often performed poorly in multiclass classification. However, we found that the convergence of the IRLS algorithm is stable and that the log likelihood is monotonic increasing when a learning rate smaller than one is adopted. Also, we propose the use of an expectation-conditional maximization (ECM) algorithm to train ME networks. Its performance is demonstrated to be superior to the IRLS algorithm on some simulated and real data sets.

The Apparitions of the Virgin Mary in Medjugorje: The convergence of Croation nationalism on her apparitions

This article concerns the alleged apparitions of the Virgin Mary in one of the most popular, 'active' apparitional sites in the world: Medjugorje in Bosnia and Herzegovina. The connection between nationalist discourse and apparitions has often been observed and noted in the literature on nationalism; however, the examples of this connection are scattered in the literature and the question why the apparitional phenomenon so easily lends itself to co-option into nationalist discourse has never been addressed. This article explores this question by showing that what binds the two phenomena together is the idea of 'chosenness' and 'specialness', which in turn can be theoretically linked to discussions about national election in the literature on nationalism. This article illustrates the convergence of nationalist and apparitional discourses by drawing on a selected number of examples of how the apparitions in Medjugorje have been appropriated by Croatian nationalist discourse.