The time dimension of neural network models

This review attempts to provide an insightful perspective on the role of time within neural network models and the use of neural networks for problems involving time. The most commonly used neural network models are defined and explained giving mention to important technical issues but avoiding great detail. The relationship between recurrent and feedforward networks is emphasised, along with the distinctions in their practical and theoretical abilities. Some practical examples are discussed to illustrate the major issues concerning the application of neural networks to data with various types of temporal structure, and finally some highlights of current research on the more difficult types of problems are presented.

The exact sample complexity of PAC-learning problems with unit VC dimension

The Vapnik-Chervonenkis (VC) dimension is a combinatorial measure of a certain class of machine learning problems, which may be used to obtain upper and lower bounds on the number of training examples needed to learn to prescribed levels of accuracy. Most of the known bounds apply to the Probably Approximately Correct (PAC) framework, which is the framework within which we work in this paper. For a learning problem with some known VC dimension, much is known about the order of growth of the sample-size requirement of the problem, as a function of the PAC parameters. The exact value of sample-size requirement is however less well-known, and depends heavily on the particular learning algorithm being used. This is a major obstacle to the practical application of the VC dimension. Hence it is important to know exactly how the sample-size requirement depends on VC dimension, and with that in mind, we describe a general algorithm for learning problems having VC dimension 1. Its sample-size requirement is minimal (as a function of the PAC parameters), and turns out to be the same for all non-trivial learning problems having VC dimension 1. While the method used cannot be naively generalised to higher VC dimension, it suggests that optimal algorithm-dependent bounds may improve substantially on current upper bounds.

Comment déterminer la valeur qualitative des indéfinis de sélection arbitraire?

The dimension qualitative of arbitrary selection undefined anyone and any what is envisaged in this article. It provides a procedure to identify jobs which are demonstrating an interpretation qualitative despite changes to which this reading is submitted. Contexts that promote and discourage this interpretation undergo an examination detailed. This gives rise to the proposal based test review on some argumentatifs linkages. Tests and contexts operate with (in) compatibility with qualitative reads the causes are envisaged. The ability to dispose of the qualitative nature of an interpretation provides advanced identification procedure constitute a new result that represents a necessary condition for an explanatory of semantics the axiology.

Distributed algorithms for global optimization on sparse networks of arbitrary bandwidths

The optimization of resource allocation in sparse networks with real variables is studied using methods of statistical physics. Efficient distributed algorithms are devised on the basis of insight gained from the analysis and are examined using numerical simulations, showing excellent performance and full agreement with the theoretical results.

Innovation and regional absorptive capacity:the labour market dimension

In 2003, Eurostat published an 'experimental' dataset on regional innovation levels derived from the Second Community Innovation Survey. This dataset, part of the European Innovation Scoreboard, also contains a range of regional labour market indicators. In this paper, we report an exploratory analysis of this data, focussing on how the labour market characteristics of regions shape regions' absorptive capacity (RACAP) and their ability to assimilate knowledge from public and externally conducted R&D. In particular, we aim to establish whether labour market aspects of RACAP are more important for innovation in prosperous or lagging regions of the European Union (EU). © Springer-Verlag 2006.

Coherent propagation and energy transfer in low-dimension nonlinear arrays

We present a theory of coherent propagation and energy or power transfer in a low-dimension array of coupled nonlinear waveguides. It is demonstrated that in the array with nonequal cores (e.g., with the central core) stable steady-state coherent multicore propagation is possible only in the nonlinear regime, with a power-controlled phase matching. The developed theory of energy or power transfer in nonlinear discrete systems is rather generic and has a range of potential applications including both high-power fiber lasers and ultrahigh-capacity optical communication systems. © 2012 American Physical Society.

Investigation of two-wave mixing in arbitrary oriented sillenite crystals

A generalized systematic description of the Two-Wave Mixing (TWM) process in sillenite crystals allowing for arbitrary orientation of the grating vector is presented. An analytical expression for the TWM gain is obtained for the special case of plane waves in a thin crystal (|g|d«1) with large optical activity (|g|/?«1, g is the coupling constant, ? the rotatory power, d the crystal thickness). Using a two-dimensional formulation the scope of the nonlinear equations describing TWM can be extended to finite beams in arbitrary geometries and to any crystal parameters. Two promising applications of this formulation are proposed. The polarization dependence of the TWM gain is used for the flattening of Gaussian beam profiles without expanding them. The dependence of the TWM gain on the interaction length is used for the determination of the crystal orientation. Experiments carried out on Bi12GeO20 crystals of a non-standard cut are in good agreement with the results of modelling.

German-Irish corporate relationships:the cultural dimension

The thesis raises the question of whether or not in an age of internationalisation and globalisation, the cultural differences which exist between Germany and Ireland are still relevant to German-Irish corporate relationships or have internationally accepted best practices removed culture from the equation? The first three chapters establish the theoretical framework of the thesis by outlining the broadly culturalist/institutionalist approach, based on the work of Hofstede and Maurice et al, to be pursued, profiling the business cultures of both countries by analysing the components of their respective national institutional frameworks, and the examining existing approaches to the study of mother company-foreign subsidiary relationships. Chapters four to seven constitute the empirical section of the thesis. Using the interviews carried out with two sample groups (Sample Group A: 15 German mother companies and 14 of their Irish operations and Sample Group B: 7 Irish mother companies and 9 of their German operations), the mother companies in both groups are examined to see whether or not they demonstrate characteristics which are in keeping with their national business cultures. Their foreign operations are then analysed as is the mother company-foreign subsidiary relationship to determine whether or not any mother company influences are visible. The general approaches adopted by the two groups of mother companies to their foreign operations are compared and contrasted. Finally, differences in national attitudes and values are identified and their impact assessed. The analysis reveals that despite existing pressures towards convergence, the cultural differences between both countries are still relevant to the relationship particularly at the level of attitudes and values and although similarities in the mother company approaches to their subsidiaries are present, national specificities may nevertheless be detected.

Reduced dimension models for weather forecasting

DUE TO COPYRIGHT RESTRICTIONS ONLY AVAILABLE FOR CONSULTATION AT ASTON UNIVERSITY LIBRARY AND INFORMATION SERVICES WITH PRIOR ARRANGEMENT

Synthesis of arbitrary group delay responses with all-pass optical cavities structures

We propose a systematic method for the synthesis of arbitrary group delay responses by using all-pass structures of coupled optical cavities. Optimum structure parameters design, in terms of filter order and accuracy, are obtained.