Real-time control of a Tokamak plasma using neural networks

This paper presents results from the first use of neural networks for the real-time feedback control of high temperature plasmas in a Tokamak fusion experiment. The Tokamak is currently the principal experimental device for research into the magnetic confinement approach to controlled fusion. In the Tokamak, hydrogen plasmas, at temperatures of up to 100 Million K, are confined by strong magnetic fields. Accurate control of the position and shape of the plasma boundary requires real-time feedback control of the magnetic field structure on a time-scale of a few tens of microseconds. Software simulations have demonstrated that a neural network approach can give significantly better performance than the linear technique currently used on most Tokamak experiments. The practical application of the neural network approach requires high-speed hardware, for which a fully parallel implementation of the multi-layer perceptron, using a hybrid of digital and analogue technology, has been developed.

EM optimization of latent-variable density models

There is currently considerable interest in developing general non-linear density models based on latent, or hidden, variables. Such models have the ability to discover the presence of a relatively small number of underlying `causes' which, acting in combination, give rise to the apparent complexity of the observed data set. Unfortunately, to train such models generally requires large computational effort. In this paper we introduce a novel latent variable algorithm which retains the general non-linear capabilities of previous models but which uses a training procedure based on the EM algorithm. We demonstrate the performance of the model on a toy problem and on data from flow diagnostics for a multi-phase oil pipeline.

Theoretical foundations of neural networks

Neural networks have often been motivated by superficial analogy with biological nervous systems. Recently, however, it has become widely recognised that the effective application of neural networks requires instead a deeper understanding of the theoretical foundations of these models. Insight into neural networks comes from a number of fields including statistical pattern recognition, computational learning theory, statistics, information geometry and statistical mechanics. As an illustration of the importance of understanding the theoretical basis for neural network models, we consider their application to the solution of multi-valued inverse problems. We show how a naive application of the standard least-squares approach can lead to very poor results, and how an appreciation of the underlying statistical goals of the modelling process allows the development of a more general and more powerful formalism which can tackle the problem of multi-modality.

Mixtures of probabilistic principal component analysers

Principal component analysis (PCA) is one of the most popular techniques for processing, compressing and visualising data, although its effectiveness is limited by its global linearity. While nonlinear variants of PCA have been proposed, an alternative paradigm is to capture data complexity by a combination of local linear PCA projections. However, conventional PCA does not correspond to a probability density, and so there is no unique way to combine PCA models. Previous attempts to formulate mixture models for PCA have therefore to some extent been ad hoc. In this paper, PCA is formulated within a maximum-likelihood framework, based on a specific form of Gaussian latent variable model. This leads to a well-defined mixture model for probabilistic principal component analysers, whose parameters can be determined using an EM algorithm. We discuss the advantages of this model in the context of clustering, density modelling and local dimensionality reduction, and we demonstrate its application to image compression and handwritten digit recognition.

Developments of the generative topographic mapping

The generative topographic mapping (GTM) model was introduced by Bishop et al. (1998, Neural Comput. 10(1), 215-234) as a probabilistic re- formulation of the self-organizing map (SOM). It offers a number of advantages compared with the standard SOM, and has already been used in a variety of applications. In this paper we report on several extensions of the GTM, including an incremental version of the EM algorithm for estimating the model parameters, the use of local subspace models, extensions to mixed discrete and continuous data, semi-linear models which permit the use of high-dimensional manifolds whilst avoiding computational intractability, Bayesian inference applied to hyper-parameters, and an alternative framework for the GTM based on Gaussian processes. All of these developments directly exploit the probabilistic structure of the GTM, thereby allowing the underlying modelling assumptions to be made explicit. They also highlight the advantages of adopting a consistent probabilistic framework for the formulation of pattern recognition algorithms.

Developing the public health function of locum pharmacists under the auspices of the new pharmacy contract

Focal Point - There are reduced opportunities for locum pharmacists to access training and education that meets their needs and enables them to play a full role under the new pharmacy contract - Eighty-six per cent of locums consider themselves to be more health professional than business person, compared to just 48% of pharmacy owners - Forty per cent of locums believe that a lack of access to training is a major barrier to the development of their public health function - While locum pharmacists are arguably more likely to embrace 'professionalising', patient-care-based roles, they are also the group least likely to be able to access the necessary training to fulfill such roles Introduction It has been suggested that locum pharmacists do not want the business-based responsibilities (e.g. staff management, meeting targets, etc) that come with pharmacy management.1 Research also suggests that locums derive great satisfaction from the health-professional aspects of the pharmacists’ role (e.g. patient contact, the provision of advice, etc).1 However, upon the introduction of the new pharmacy contract (April 2005), concerns were expressed that it was becoming increasingly difficult for locum pharmacists to access training and education that would meet their needs and enable them to play a full role under the new framework.2,3 Method After piloting, in August 2006 a self-completion postal questionnaire was sent to a random sample of practising community pharmacists, stratified for country and sex, within Great Britain (n = 1998), with a follow-up to non-responders 4 weeks later. Data were analysed using SPSS (v12.0). A final response rate of 51% (n = 1023/1998) was achieved. Respondents were asked ‘indicate how you view yourself as a pharmacist’ – in terms of their relative focus on the health-professional and business aspects of their role. Respondents were also asked ‘do you consider a lack of training opportunities to be a barrier to the development of the public health role of community pharmacists?’. Results Locums were significantly more likely than owners or employees to consider each factor a major barrier. Discussion Four in 10 locums consider a lack of training opportunities to constitute a major barrier to the development of their public health function. Pharmacy may not be able to provide the services required of it by the policy agenda if pharmacists are unable to be involved in extended role activities through a lack of training opportunities. Therefore, the paradox that needs to be addressed is that while locum pharmacists are arguably more likely to embrace ‘professionalising’, patient-care-based roles, they are also the group least likely to be able to access training to fulfil such roles. The training needs of this large subset of the pharmacist population need to be assessed and met if the whole community pharmacy workforce is going to maximise its contribution to public health under the new contractual framework. References 1 Shann P, Hassell K. An exploration of the diversity and complexity of the pharmacy locum workforce. London: Royal Pharmaceutical Society of Great Britain; 2004. 2 Almond M. Locums – key players in workforce – cast adrift as contract launched. Pharm J 2005;274:420. 3 Bishop DH. A lack of appreciation of what really happens. Pharm J 2005;274:451.

Privatisation, corporate control and employment growth:evidence from a panel of large Polish firms, 1996-2002

Using data on Polish firms this paper examines the relationship between corporate control structures, sales growth and the determinants of employment change. Privatised and de novo firms are the main drivers of employment growth; in the case of de novo firms, it is foreign ownership which underpins the result. Being privatised has a positive impact on employment but this is concentrated within a range of 3–6 years after privatisation. There are no systematic differences in employment response to negative sales growth across the ownership categories. Employment in state firms is less responsive to positive sales growth. From these results we infer that the behaviour of state firms is affected by both insider rent sharing and binding budget constraints.

Privatisation, corporate control and employment growth:evidence from a panel of large Polish firms, 1996-2002

Using panel data on large Polish firms this paper examines the relationship between corporate control structures, sales growth and the determinants of employment change during the period 1996-2002. We find that privatised and de novo firms are the main drivers of employment growth and that, in the case of de novo firms, it is foreign ownership which underpins the result. Interestingly, we find that being privatised has a positive impact on employment growth but that this impact is concentrated within a range of three to six years after privatisation. In contrast with the findings of earlier literature, we find evidence that there are no systematic differences in employment response to negative sales growth across the ownership categories. On the other hand, employment in state firms is less responsive to positive sales growth. From these combined results we infer that the behaviour of state firms is constrained by both insider rent sharing and binding budget constraints. Consistent with this, we find that privatised companies, three to six years post-privatisation, are the firms for whom employment is most responsive to positive sales growth and as such, offer the best hope for rapid labour market expansion.

Wage and employment decisions of enterprises in downsized industries:deliverable No. 11 to 13

The 5th framework programme research project ACCESSLAB researches the capability of candidate countries’ regions to deal with asymmetric shocks. Its goal is to provide analysts and policy makers with research results relevant to the process of enlargement. The project takes a broad and comparative view of labour market adjustments to address these issues. It examines the topic from both a macroeconomic and microeconomic viewpoint. It considers different adjustment mechanisms in depth and compares results with the European Union. It draws on a) the experiences in transition countries in the last decade, b) the experience of German integration and c) the experiences of border regions to gain insights on the likely regional labour market effects of accession of the candidate countries.

Brain, mind and consciousness in the history of neuroscience

From Platonic and Galenic roots, the first well developed ventricular theory of brain function is due to Bishop Nemesius, fourth century C.E. Although more interested in the Christian concept of soul, St. Augustine, too addressed the question of the location of the soul, a problem that has endured in various guises to the present day. Other notable contributions to ventricular psychology are the ninth century C.E. Arabic writer, Qusta ibn Lūqā, and an early European medical text written by the twelfth century C.E. author, Nicolai the Physician. By the time of Albertus Magnus, so-called medieval cell doctrine was a well-developed model of brain function. By the sixteenth century, Vesalius no longer understands the ventricles to be imaginary cavities designed to provide a physical basis for faculty psychology but as fluid-filled spaces in the brain whose function is yet to be determined