Topographic mappings and feed-forward neural networks

This thesis is a study of the generation of topographic mappings - dimension reducing transformations of data that preserve some element of geometric structure - with feed-forward neural networks. As an alternative to established methods, a transformational variant of Sammon's method is proposed, where the projection is effected by a radial basis function neural network. This approach is related to the statistical field of multidimensional scaling, and from that the concept of a 'subjective metric' is defined, which permits the exploitation of additional prior knowledge concerning the data in the mapping process. This then enables the generation of more appropriate feature spaces for the purposes of enhanced visualisation or subsequent classification. A comparison with established methods for feature extraction is given for data taken from the 1992 Research Assessment Exercise for higher educational institutions in the United Kingdom. This is a difficult high-dimensional dataset, and illustrates well the benefit of the new topographic technique. A generalisation of the proposed model is considered for implementation of the classical multidimensional scaling (¸mds}) routine. This is related to Oja's principal subspace neural network, whose learning rule is shown to descend the error surface of the proposed ¸mds model. Some of the technical issues concerning the design and training of topographic neural networks are investigated. It is shown that neural network models can be less sensitive to entrapment in the sub-optimal global minima that badly affect the standard Sammon algorithm, and tend to exhibit good generalisation as a result of implicit weight decay in the training process. It is further argued that for ideal structure retention, the network transformation should be perfectly smooth for all inter-data directions in input space. Finally, there is a critique of optimisation techniques for topographic mappings, and a new training algorithm is proposed. A convergence proof is given, and the method is shown to produce lower-error mappings more rapidly than previous algorithms.

Understanding policy convergence in Britain and Germany: towards a framework for analysis

This introduction considers reasons why public policies might be expected to converge between Britain and Germany, arguing that the inter-related forces of globalisation, Europeanisation, policy transfer (in various guises) and the election of centre-left governance in 1997 and 1998 could be expected to lead to such convergence. It then outlines important reasons why such convergence may not occur, due to the radically different institutional settings, as well as 'path dependence' and the resilience of established institutions all playing a role in continuing divergence in a number of important areas of public policy.

‘O convergence, where art thou?’ Regional growth and industrialization in Italy

This paper shows that the Italian economy has two long-run equilibria, which are due to the different level of industrialization between the centre-north and the south of the country. These equilibria converge until 1971 but diverge afterwards; the end of the convergence process coincides with the slowing down of Italy's industrialization policy in the South. In this paper we argue that to address this problem effectively, an economic policy completely different from that in place in needed. However, such a policy is unlikely to be implemented given the scarcity of resources and the short run nature of the political cycle.

Hebbian and homeostatic plasticity mechanisms in regular spiking and intrinsic bursting cells of cortical layer 5

Layer 5 contains the major projection neurons of the neocortex and is composed of two major cell types: regular spiking (RS) cells, which have cortico-cortical projections, and intrinsic bursting cells (IB), which have subcortical projections. Little is known about the plasticity processes and specifically the molecular mechanisms by which these two cell classes develop and maintain their unique integrative properties. In this study, we find that RS and IB cells show fundementally different experience-dependent plasticity processes and integrate Hebbian and homeostatic components of plasticity differently. Both RS and IB cells showed TNFα-dependent homeostatic plasticity in response to sensory deprivation, but IB cells were capable of a much faster synaptic depression and homeostatic rebound than RS cells. Only IB cells showed input-specific potentiation that depended on CaMKII autophosphorylation. Our findings demonstrate that plasticity mechanisms are not uniform within the neocortex, even within a cortical layer, but are specialized within subcircuits.

Time complexity and convergence analysis of domain theoretic Picard method

We present an implementation of the domain-theoretic Picard method for solving initial value problems (IVPs) introduced by Edalat and Pattinson [1]. Compared to Edalat and Pattinson's implementation, our algorithm uses a more efficient arithmetic based on an arbitrary precision floating-point library. Despite the additional overestimations due to floating-point rounding, we obtain a similar bound on the convergence rate of the produced approximations. Moreover, our convergence analysis is detailed enough to allow a static optimisation in the growth of the precision used in successive Picard iterations. Such optimisation greatly improves the efficiency of the solving process. Although a similar optimisation could be performed dynamically without our analysis, a static one gives us a significant advantage: we are able to predict the time it will take the solver to obtain an approximation of a certain (arbitrarily high) quality.

Neuronal convergence in early contrast vision: binocular summation is followed by response nonlinearity and area summation

We assessed summation of contrast across eyes and area at detection threshold ( C t). Stimuli were sine-wave gratings (2.5 c/deg) spatially modulated by cosine- and anticosine-phase raised plaids (0.5 c/deg components oriented at ±45°). When presented dichoptically the signal regions were interdigitated across eyes but produced a smooth continuous grating following their linear binocular sum. The average summation ratio ( C t1/([ C t1+2]) for this stimulus pair was 1.64 (4.3 dB). This was only slightly less than the binocular summation found for the same patch type presented to both eyes, and the area summation found for the two different patch types presented to the same eye. We considered 192 model architectures containing each of the following four elements in all possible orders: (i) linear summation or a MAX operator across eyes, (ii) linear summation or a MAX operator across area, (iii) linear or accelerating contrast transduction, and (iv) additive Gaussian, stochastic noise. Formal equivalences reduced this to 62 different models. The most successful four-element model was: linear summation across eyes followed by nonlinear contrast transduction, linear summation across area, and late noise. Model performance was enhanced when additional nonlinearities were placed before binocular summation and after area summation. The implications for models of probability summation and uncertainty are discussed.

Determinants of sales performance in insurance industry:a cross-cultural comparison between the UK and Malaysia

This thesis is concerned with an empirical investigation of the factors that predict a successful salesperson, using a cross-cultural comparison of two countries: the UK and Malaysia. Besides collecting quantitative data, qualitative data on organisational, environmental and cultural factors were also collected through interviews, personal and case observations. The quantitative data consist of sixteen independent factors and three dependent factors. The independent variables include self-efficacy, self-esteem, locus of control, self-monitoring, extrinsic motivation, intrinsic motivation, experience, training perception, role ambiguity, role conflict, role inaccuracy, gender, age, education, race and religion. The dependent variables are performance target achieved, performance earnings and performance ratings. Questionnaires were distributed to about 500 salespersons in each country, from three insurance companies in the UK and two insurance companies in Malaysia. Response rates were 75 and 50 percent from the UK and Malaysia respectively. The survey results indicated that a salesperson's performance in the UK is predicted by self-efficacy, internal locus of control, self-esteem, extrinsic motivation, experience, training perceptions, role conflict and gender. In Malaysia, a salesperson's performance is predicted by self-efficacy, self-monitoring, experience, role conflict, role ambiguity, education, gender, race and religion. Self-efficacy, experience, role conflict and gender are common predictors of salespersons' performance in both cultures. The likely explanation for these results is culture differences, i.e. UK has a homogeneous culture, while Malaysia has a heterogeneous one. Results from the case observations, such as organisational and environmental factors, give supporting evidence in explaining the empirical results. Implications from the findings are discussed from two aspects: (1) theoretical implications for divergence/convergence theory, Hofstede's model, Churchill's model, and (2) managerial implications for selection, training, motivation and appraisal.

Exploiting uncertainty in nonlinear stochastic control problem

This work introduces a novel inversion-based neurocontroller for solving control problems involving uncertain nonlinear systems which could also compensate for multi-valued systems. The approach uses recent developments in neural networks, especially in the context of modelling statistical distributions, which are applied to forward and inverse plant models. Provided that certain conditions are met, an estimate of the intrinsic uncertainty for the outputs of neural networks can be obtained using the statistical properties of networks. More generally, multicomponent distributions can be modelled by the mixture density network. Based on importance sampling from these distributions a novel robust inverse control approach is obtained. This importance sampling provides a structured and principled approach to constrain the complexity of the search space for the ideal control law. The developed methodology circumvents the dynamic programming problem by using the predicted neural network uncertainty to localise the possible control solutions to consider. Convergence of the output error for the proposed control method is verified by using a Lyapunov function. Several simulation examples are provided to demonstrate the efficiency of the developed control method. The manner in which such a method is extended to nonlinear multi-variable systems with different delays between the input-output pairs is considered and demonstrated through simulation examples.

Constructing institutional change: emergence, contestation and convergence of business models in the field of Java application servers

How are innovative new business models established if organizations constantly compare themselves against existing criteria and expectations? The objective is to address this question from the perspective of innovators and their ability to redefine established expectations and evaluation criteria. The research questions ask whether there are discernible patterns of discursive action through which innovators theorize institutional change and what role such theorizations play for mobilizing support and realizing change projects. These questions are investigated through a case study on a critical area of enterprise computing software, Java application servers. In the present case, business practices and models were already well established among incumbents with critical market areas allocated to few dominant firms. Fringe players started experimenting with a new business approach of selling services around freely available opensource application servers. While most new players struggled, one new entrant succeeded in leading incumbents to adopt and compete on the new model. The case demonstrates that innovative and substantially new models and practices are established in organizational fields when innovators are able to refine expectations and evaluation criteria within an organisational field. The study addresses the theoretical paradox of embedded agency. Actors who are embedded in prevailing institutional logics and structures find it hard to perceive potentially disruptive opportunities that fall outside existing ways of doing things. Changing prevailing institutional logics and structures requires strategic and institutional work aimed at overcoming barriers to innovation. The study addresses this problem through the lens of (new) institutional theory. This discourse methodology traces the process through which innovators were able to establish a new social and business model in the field.

Iterative subspace analysis based on feature line distance

Nearest feature line-based subspace analysis is first proposed in this paper. Compared with conventional methods, the newly proposed one brings better generalization performance and incremental analysis. The projection point and feature line distance are expressed as a function of a subspace, which is obtained by minimizing the mean square feature line distance. Moreover, by adopting stochastic approximation rule to minimize the objective function in a gradient manner, the new method can be performed in an incremental mode, which makes it working well upon future data. Experimental results on the FERET face database and the UCI satellite image database demonstrate the effectiveness.

Subspace modelling for structured noise suppression

The problem of structured noise suppression is addressed by i)modelling the subspaces hosting the components of the signal conveying the information and ii)applying a nonlin- ear non-extensive technique for effecting the right separation. Although the approach is applicable to all situations satisfying the hypothesis of the proposed framework, this work is motivated by a particular scenario, namely, the cancellation of low frequency noise in broadband seismic signals.

New strategy to overcome the intrinsic difficulty of phospholipids solubilisation and delivery to the eye

Purpose: Lipids play a vital role at interfaces such as the tear film in the protection of the anterior eye. Their role is to act as lubricants and reduce surface and interfacial tension. Although there is a lack of appropriate methods to solubilize and dilute phospholipids to the tear film. Here, we report that styrene-maleic acid copolymers (PSMA), can form polymer–lipid complexes in the form of monodisperse nanometric particles, which can easily solubilise these phospholipid molecules by avoiding for example, the use of any kind of surfactant. Method: The interactions of PSMA with phospholipids have been studied by its adsorption from aqueous solutions into monolayers of dimyristoyl-phosphorylcholine (DMPC). The Langmuir trough (LT) technique is used to study this pH-dependant complex formation. The formed nanoparticles have been also analysed by 31P NMR, particle size distribution by light scattering (DLS) and morphology by electron microscopy (SEM). Results: The LT has been found to be a useful technique for in vitro simulation of in vivo lipid layer behaviour: The limiting surface pressure of unstable tear films ranges between 20 and 30 mN/m. More stable tear films show an increase in surface pressure, within the range of 35–45 mN/m. The DMPC monolayers have a limiting surface pressure of 38 mN/m (water), and 45 mN/m (pH 4 buffer), and the PSMA-DMPC complexes formed at pH 4 have a value of 42 mN/m, which resembles that of the stable tear film. The average particle size distribution is 53 ± 10 nm with a low polydispersity index (PDI) of 0.24 ± 0.03. Conclusions: New biocompatible and cheap lipid solubilising agents such as PSMA can be used for the study of the tear film composition and properties. These polymer–lipid complexes in the form of nanoparticles can be used to solubilise and release in a controlled way other hydrophobic molecules such as some drugs or proteins.