Thesis (M.S.)--University of Wisconsin--Madison. / Typescript. Includes bibliographical references.

Thesis (M.S.)--University of Wisconsin--Madison. / Typescript. Title from title screen (viewed Oct. 1, 2007). Includes bibliographical references. Online version of the print original.

No description available.

Mémoire (M.Ed.)--Université du Québec à Chicoutimi, 1999. / Document électronique également accessible en format PDF. CaQCU

No description available.

La Terre et l'Univers font l'objet d'une mise en discours dans la société et d'une couverture médiatique sans précédent. L'astrophysicien et vulgarisateur scientiliquè Hubert Reeves participe de cette mise en discours en humanisant la planète et l'Univers afin de nous faire prendre conscience de l'urgence de la situation environnementale. Nous formulons l'hypothèse que son discours de vulgarisation fonctionne comme un discours religieux en ceci qu'il fonde les valeurs écologiques par la ré-humanisation de la nature et l'humanisation de l'homme par l'homme. Discours qui établit un nouvel éthos et l'ancre dans une cosmologie où l'être humain a une place et un dessein. Il espère que l'homme prendra soin de l'univers en conférant à ce dernier un statut ontologique supérieur à l'humanité. Hubert Reeves procède à l'élaboration d'une nouvelle vision du monde fondée sur l'emploi de métaphores et faisant continuellement référence à des univers symboliques (cosmos, univers, mort). Dans l'univers sécularisé qu'est le nôtre, Hubert Reeves prend le relais de la Bible dans le monde judéo-chrétien en expliquant l'univers qui nous entoure, en lui donnant un sens, en expliquant la vie, en racontant la marche du monde, en émerveillant... À une époque où le discours techno-scientifique tend à banaliser l'univers. Hubert Reeves le ré-humanise, le re-symbolise en employant des métaphores pour redonner sens aux discours écologique et scientifique. Il est dans l'urgence. On doit retourner aux fondements de l'univers pour réinscrire notre désir de poursuivre notre avenir collectif sur la planète bleue. Cette permanence de l'humanité n'est pas garantie par notre historicité mais bien par un changement d'éthos collectif par rapport à la nature. ______________________________________________________________________________ MOTS-CLÉS DE L’AUTEUR : Science, Religion, Hubert Reeves, Musique classique, Cosmologie, Éthique, Métaphore, Urgence.

This literary/historical novel details the life of African-American Deputy US Marshal Bass Reeves between the years 1838-1862 and 1883-1884. One plotline depicts Reeves’s youth as a slave, including his service as a body servant to a Confederate cavalry officer during the Civil War. Another plotline depicts him years later, after Emancipation, at the height of his deputy career, when he has become the most feared, most successful lawman in Indian Territory, the largest federal jurisdiction in American history and the most dangerous part of the Old West. A preface explores the uniqueness of this project’s historical relevance and literary positioning as a neo-slave narrative, and addresses a few liberties that I take with the historical record.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

No description available.

A new neural network approach for performing matrix computations is presented. The idea of this approach is to construct a feed-forward neural network (FNN) and then train it by matching a desired set of patterns. The solution of the problem is the converged weight of the FNN. Accordingly, unlike the conventional FNN research that concentrates on external properties (mappings) of the networks, this study concentrates on the internal properties (weights) of the network. The present network is linear and its weights are usually strongly constrained; hence, complicated overlapped network needs to be construct. It should be noticed, however, that the present approach depends highly on the training algorithm of the FNN. Unfortunately, the available training methods; such as, the original Back-propagation (BP) algorithm, encounter many deficiencies when applied to matrix algebra problems; e. g., slow convergence due to improper choice of learning rates (LR). Thus, this study will focus on the development of new efficient and accurate FNN training methods. One improvement suggested to alleviate the problem of LR choice is the use of a line search with steepest descent method; namely, bracketing with golden section method. This provides an optimal LR as training progresses. Another improvement proposed in this study is the use of conjugate gradient (CG) methods to speed up the training process of the neural network. The computational feasibility of these methods is assessed on two matrix problems; namely, the LU-decomposition of both band and square ill-conditioned unsymmetric matrices and the inversion of square ill-conditioned unsymmetric matrices. In this study, two performance indexes have been considered; namely, learning speed and convergence accuracy. Extensive computer simulations have been carried out using the following training methods: steepest descent with line search (SDLS) method, conventional back propagation (BP) algorithm, and conjugate gradient (CG) methods; specifically, Fletcher Reeves conjugate gradient (CGFR) method and Polak Ribiere conjugate gradient (CGPR) method. The performance comparisons between these minimization methods have demonstrated that the CG training methods give better convergence accuracy and are by far the superior with respect to learning time; they offer speed-ups of anything between 3 and 4 over SDLS depending on the severity of the error goal chosen and the size of the problem. Furthermore, when using Powell's restart criteria with the CG methods, the problem of wrong convergence directions usually encountered in pure CG learning methods is alleviated. In general, CG methods with restarts have shown the best performance among all other methods in training the FNN for LU-decomposition and matrix inversion. Consequently, it is concluded that CG methods are good candidates for training FNN of matrix computations, in particular, Polak-Ribidre conjugate gradient method with Powell's restart criteria.