Abstract. We describe two important sets of tools for neural network modeling: prediction risk estimation and network architecture selection. Prediction risk is defined as the expected performance of an estimator in predicting new observations. Estimated prediction risk can be used both for estimating the quality of model predictions and for model selection. Prediction risk estimation and model selection are especially important for problems with limited data. Techniques for estimating prediction risk include data resampling algorithms such as nonlinear cross–validation (NCV) and algebraic formulae such as the predicted squared error (PSE) and generalized prediction error (GPE). We show that exhaustive search over the space of network architectures is computationally infeasible even for networks of modest size. This motivates the use of heuristic strategies that dramatically reduce the search complexity. These strategies employ directed search algorithms, such as selecting the number of nodes via sequential network construction (SNC) and pruning inputs and weights via sensitivity based pruning (SBP) and optimal brain damage (OBD) respectively.

Generalizations of Roo's score test are receiving increased attention, especially in the econometrics and biostatistics literature. These generalizations are able to account for certain model inadequacies or lack of knowledge by use of empirical variance estimates. This article shows how the various forms of the generalized test statistic arise from Taylor expansion of the estimating equations. The general estimating equations structure unifies a variety of applications and helps suggest new areas of application. Key Words: composite null hypothesis, empirical variance, estimating equations, information sandwich, Lagrange multiplier test, misspecified likelihood, observed information, robust inference. ON GENERALIZED SCORE TESTS 1.

: In this paper, we apply the synchronous approach to real time active visual reconstruction. It illustrates the adequateness of Signal, a synchronous data flow programming language and environment, for the specification of a system dealing with various domains such as robot control, computer vision and programmation of hierarchical parallel automaton. More precisely, one application consists in the 3D structure estimation of a set of geometrical primitives using an active vision paradigm. At the level of camera motion control, the visual servoing approach (a closed loop with respect to vision data) is specified and implemented in Signal as a function from sensor inputs to control outputs. Furthermore, the 3D reconstruction method is based on the "structure from controlled motion" approach (constraining camera motion for optimal estimation). Its specification is made in parallel to visual servoing, and involves the delay mechanism of Signal for the specification of filters. This recons...

This paper deals with the problem of 3D structure estimation of a set of geometrical primitives in an active vision context. Our method is based on the structure from controlled motion approach which consists in constraining the camera motion in order to obtain an optimal estimation of the 3D structure of a geometrical primitive. Since this approach involves to focus on the considered primitive, wc present a method for connecting up many estimations. It is centered on visual 2D information and rough estimation of the structure of the studied primitives. It 11ows us to have an optimal estimation of each primitive of a scene composed of cyhndcrs, segments and polygons. Our method has bccn implemented with an automata network using the real time SIGNAL language. These automata arc able to connect up the different stages of the reconstruction process: selection, focusing, optimal estimation of the selected primitive and, simultaneously, rough estimation of the other ones. Rcl time experiments on a robotic arc presented and experimental results arc given.

eingereicht an der

For a Web download or e-book: Your use of this publication shall be governed by the terms established by the vendor at the time you acquire this publication. U.S. Government Restricted Rights Notice: Use, duplication, or disclosure of this software and related documentation by the U.S. government is subject to the Agreement with SAS Institute and the restrictions set forth in FAR 52.227-19, Commercial Computer Software-Restricted Rights (June 1987).

For a Web download or e-book: Your use of this publication shall be governed by the terms established by the vendor at the time you acquire this publication. The scanning, uploading, and distribution of this book via the Internet or any other means without the permission of the publisher is illegal and punishable by law. Please purchase only authorized electronic editions and do not participate in or encourage electronic piracy of copyrighted materials. Your support of others ’ rights is appreciated. U.S. Government Restricted Rights Notice: Use, duplication, or disclosure of this software and related documentation by the

Supporting student success well in a global economy under the very realworld constraints of scarce resources requires certain economies of instructional practice. Teachers with ever-more-limited time and ever-increasing demands need to make critical decisions concerning educational interventions for individual

For a Web download or e-book: Your use of this publication shall be governed by the terms established by the vendor at the time you acquire this publication. The scanning, uploading, and distribution of this book via the Internet or any other means without the permission of the publisher is illegal and punishable by law. Please purchase only authorized electronic editions and do not participate in or encourage electronic piracy of copyrighted materials. Your support of others ’ rights is appreciated. U.S. Government Restricted Rights Notice: Use, duplication, or disclosure of this software and related documentation by the