Regularisation of mixture density networks

Mixture Density Networks are a principled method to model conditional probability density functions which are non-Gaussian. This is achieved by modelling the conditional distribution for each pattern with a Gaussian Mixture Model for which the parameters are generated by a neural network. This thesis presents a novel method to introduce regularisation in this context for the special case where the mean and variance of the spherical Gaussian Kernels in the mixtures are fixed to predetermined values. Guidelines for how these parameters can be initialised are given, and it is shown how to apply the evidence framework to mixture density networks to achieve regularisation. This also provides an objective stopping criteria that can replace the `early stopping' methods that have previously been used. If the neural network used is an RBF network with fixed centres this opens up new opportunities for improved initialisation of the network weights, which are exploited to start training relatively close to the optimum. The new method is demonstrated on two data sets. The first is a simple synthetic data set while the second is a real life data set, namely satellite scatterometer data used to infer the wind speed and wind direction near the ocean surface. For both data sets the regularisation method performs well in comparison with earlier published results. Ideas on how the constraint on the kernels may be relaxed to allow fully adaptable kernels are presented.

Regularisation of mixture density networks

Mixture Density Networks are a principled method to model conditional probability density functions which are non-Gaussian. This is achieved by modelling the conditional distribution for each pattern with a Gaussian Mixture Model for which the parameters are generated by a neural network. This thesis presents a novel method to introduce regularisation in this context for the special case where the mean and variance of the spherical Gaussian Kernels in the mixtures are fixed to predetermined values. Guidelines for how these parameters can be initialised are given, and it is shown how to apply the evidence framework to mixture density networks to achieve regularisation. This also provides an objective stopping criteria that can replace the `early stopping' methods that have previously been used. If the neural network used is an RBF network with fixed centres this opens up new opportunities for improved initialisation of the network weights, which are exploited to start training relatively close to the optimum. The new method is demonstrated on two data sets. The first is a simple synthetic data set while the second is a real life data set, namely satellite scatterometer data used to infer the wind speed and wind direction near the ocean surface. For both data sets the regularisation method performs well in comparison with earlier published results. Ideas on how the constraint on the kernels may be relaxed to allow fully adaptable kernels are presented.

Dissipative nonlinear structures in fiber optics

Optical fiber materials exhibit a nonlinear response to strong electric fields, such as those of optical signals confined within the small fiber core. Fiber nonlinearity is an essential component in the design of the next generation of advanced optical communication systems, but its use is often avoided by engineers because of its intractability. The application of nonlinear technologies in fiber optics offers new opportunities for the design of photonic systems and devices. In this chapter, we make an overview of recent progress in mathematical theory and practical applications of temporal dissipative solitons and self-similar nonlinear structures in optical fiber systems. The design of all-optical high-speed signal processing devices, based on nonlinear dissipative structures, is discussed.

RAMPs as drug targets

The receptor activity-modifying protein (RAMP) family of membrane proteins regulates G protein-coupled receptor (GPCR) function in several ways. RAMPs can alter their pharmacology and signalling as well as the trafficking of these receptors to and from the cell surface. Accordingly, RAMPs may be exploited as drug targets, offering new opportunities for regulating the function of therapeutically relevant RAMP-interacting GPCRs. For example, several small molecule antagonists of RAMP1/ calcitonin receptor-like receptor complexes, which block the actions of the neuropeptide calcitonin gene-related peptide are in development for the treatment of migraine headache.

Building a strategy toolkit:lessons from business : executive briefings

In an increasingly competitive business environment, the ability to spot and seize new opportunities, to plot a path of successful growth for an organisation, and to use resources effectively and efficiently, becomes paramount. Managers have a number of management tools at their disposal to help meet the challenges that they face. By consulting with both business academics and alumni on their knowledge and use of strategy tools, this report contains a number of strategy tools that managers would benefit from being familiar and using in their work, particularly as their experience and seniority increases.

Disentangling Double Politikverflechtung? The implications of the federal reforms for Bund-Länder relations on Europe

The recent reforms of German federalism (Reform I) have established a new framework for Bund–Länder co-operation on EU policy. These seek to safeguard Germany's ability to co-operate in Europe by disentangling the joint roles and responsibilities bound up within the complex arrangements of the EU policy-making system, defined as a multiple framework of joint decisions, or doppelte Politikverflechtung. Whilst on the surface, the reforms enacted may be read as a success for the Länder in their bid to secure autonomy on European issues, closer analysis reveals that these changes may in fact hamper the Länder agenda on European issues, closing off new opportunities for influence.

The Library of Birmingham project:lifelong learning for the digital age

The Library of Birmingham (LoB) is a £193million project designed to provide a new space for lifelong learning and knowledge growth, a physical and virtual portal for Birmingham's citizens to the wider world. In cooperation with a range of private, public, and third-sector bodies, as well as individual citizens, the library, due to open in June 2013, will articulate a continuing process of organic growth and emergence. Key delivery themes focus on: arts and creativity, citizenship and community, enterprise and innovation, learning and skills and the new media ecology. A landmark design in the heart of the cultural district of the city, the LoB aims to stimulate sustainable economic growth, urban regeneration and social inclusion by offering a wide range of new digital learning services, real and virtual community spaces, and new opportunities for interpreting and exploiting internationally significant collections of documentary archives, photography, moving image, and rare printed books. Additionally, the LoB will offer physical space for creative, cultural, enterprise, and knowledge development. This paper outlines the cultural and educational thinking that informs the project and the challenges experienced in developing innovative service redesign.

Dissipative nonlinear structures in fiber optics

Optical fiber materials exhibit a nonlinear response to strong electric fields, such as those of optical signals confined within the small fiber core. Fiber nonlinearity is an essential component in the design of the next generation of advanced optical communication systems, but its use is often avoided by engineers because of its intractability. The application of nonlinear technologies in fiber optics offers new opportunities for the design of photonic systems and devices. In this chapter, we make an overview of recent progress in mathematical theory and practical applications of temporal dissipative solitons and self-similar nonlinear structures in optical fiber systems. The design of all-optical high-speed signal processing devices, based on nonlinear dissipative structures, is discussed.

Technological enhancements to optometric clinical tests

A sizeable amount of the testing in eye care, requires either the identification of targets such as letters to assess functional vision, or the subjective evaluation of imagery by an examiner. Computers can render a variety of different targets on their monitors and can be used to store and analyse ophthalmic images. However, existing computing hardware tends to be large, screen resolutions are often too low, and objective assessments of ophthalmic images unreliable. Recent advances in mobile computing hardware and computer-vision systems can be used to enhance clinical testing in optometry. High resolution touch screens embedded in mobile devices, can render targets at a wide variety of distances and can be used to record and respond to patient responses, automating testing methods. This has opened up new opportunities in computerised near vision testing. Equally, new image processing techniques can be used to increase the validity and reliability of objective computer vision systems. Three novel apps for assessing reading speed, contrast sensitivity and amplitude of accommodation were created by the author to demonstrate the potential of mobile computing to enhance clinical measurement. The reading speed app could present sentences effectively, control illumination and automate the testing procedure for reading speed assessment. Meanwhile the contrast sensitivity app made use of a bit stealing technique and swept frequency target, to rapidly assess a patient’s full contrast sensitivity function at both near and far distances. Finally, customised electronic hardware was created and interfaced to an app on a smartphone device to allow free space amplitude of accommodation measurement. A new geometrical model of the tear film and a ray tracing simulation of a Placido disc topographer were produced to provide insights on the effect of tear film breakdown on ophthalmic images. Furthermore, a new computer vision system, that used a novel eye-lash segmentation technique, was created to demonstrate the potential of computer vision systems for the clinical assessment of tear stability. Studies undertaken by the author to assess the validity and repeatability of the novel apps, found that their repeatability was comparable to, or better, than existing clinical methods for reading speed and contrast sensitivity assessment. Furthermore, the apps offered reduced examination times in comparison to their paper based equivalents. The reading speed and amplitude of accommodation apps correlated highly with existing methods of assessment supporting their validity. Their still remains questions over the validity of using a swept frequency sine-wave target to assess patient’s contrast sensitivity functions as no clinical test provides the range of spatial frequencies and contrasts, nor equivalent assessment at distance and near. A validation study of the new computer vision system found that the authors tear metric correlated better with existing subjective measures of tear film stability than those of a competing computer-vision system. However, repeatability was poor in comparison to the subjective measures due to eye lash interference. The new mobile apps, computer vision system, and studies outlined in this thesis provide further insight into the potential of applying mobile and image processing technology to enhance clinical testing by eye care professionals.

Creating research collaboration among the global community of sales scholars:key takeaways from the 2013 AMA faculty consortium

Within the contemporary business milieu, the discipline of selling and sales management has taken on a more prominent role in recent years. Myriad factors have contributed to the rise of interest in sales including globalization, technology, more sophisticated analytical approaches and new opportunities for co-creation of value between organizations and their customers. Over the past three decades, seven faculty consortia in sales have served as milestones to document the progress 2of the field, particularly the evolution of academic research. This article provides key takeaways from the most recent American Marketing Association (AMA) Faculty Consortium in Selling and Sales Management, which had the overarching goal of fostering new opportunities for building intercontinental research teams to effectively address the substantive issues for the future of the field. © 2014 Pi Sigma Epsilon National Educational Foundation.

Fundamental Pd0/PdII redox steps in cross-coupling reactions:homogeneous, hybrid homogeneous-heterogeneous to heterogeneous mechanistic pathways for C-C couplings

C–C bond-forming, cross-coupling reactions of organohalides with nucleophilic compounds, catalysed by palladium, are amongst the most important chemical reactions available to the synthetic chemist. The intimate mechanisms of these reactions, involving Pd0/PdII redox steps, have been of great historical interest and continue to be so. The myriad of possible mechanisms is reviewed in this chapter. The interplay of mononuclear Pd species with higher order Pd species, e.g. nanoclusters/nanoparticles are considered as being equally important in cross-coupling reaction mechanisms. A focus is placed on trichotomic behaviour of cross-coupling catalytic manifolds, from homogeneous to hybrid homogeneous–heterogeneous to truly heterogeneous behaviour. For the latter, surface chemistry and metal atom leaching (and various experimental techniques) are broadly discussed. It is now clear that mechanism for general cross‐coupling reactions, that is as presented to undergraduate students studying Chemistry degrees across the world, is undoubtedly more complex than first thought. New opportunities for catalyst design have therefore emerged in the area of Pd nanoparticles and nanocatalysis, with some wonderful applications especially in chemical biology, providing a snapshot of what the future might hold.

Multi-frequency segmental bio-impedance device:design, development and applications

Bio-impedance analysis (BIA) provides a rapid, non-invasive technique for body composition estimation. BIA offers a convenient alternative to standard techniques such as MRI, CT scan or DEXA scan for selected types of body composition analysis. The accuracy of BIA is limited because it is an indirect method of composition analysis. It relies on linear relationships between measured impedance and morphological parameters such as height and weight to derive estimates. To overcome these underlying limitations of BIA, a multi-frequency segmental bio-impedance device was constructed through a series of iterative enhancements and improvements of existing BIA instrumentation. Key features of the design included an easy to construct current-source and compact PCB design. The final device was trialled with 22 human volunteers and measured impedance was compared against body composition estimates obtained by DEXA scan. This enabled the development of newer techniques to make BIA predictions. To add a ‘visual aspect’ to BIA, volunteers were scanned in 3D using an inexpensive scattered light gadget (Xbox Kinect controller) and 3D volumes of their limbs were compared with BIA measurements to further improve BIA predictions. A three-stage digital filtering scheme was also implemented to enable extraction of heart-rate data from recorded bio-electrical signals. Additionally modifications have been introduced to measure change in bio-impedance with motion, this could be adapted to further improve accuracy and veracity for limb composition analysis. The findings in this thesis aim to give new direction to the prediction of body composition using BIA. The design development and refinement applied to BIA in this research programme suggest new opportunities to enhance the accuracy and clinical utility of BIA for the prediction of body composition analysis. In particular, the use of bio-impedance to predict limb volumes which would provide an additional metric for body composition measurement and help distinguish between fat and muscle content.

Hydrothermal saline promoted grafting:a route to sulfonic acid SBA-15 silica with ultra-high acid site loading for biodiesel synthesis

A simple grafting protocol is reported which affords a ten-fold enhancement in acid site density of mesoporous sulfonic acid silicas compared to conventional syntheses, offering improved process efficiency and new opportunities for tailored supported solid acids in sustainable chemistry. This journal is

Living well to the end:a phenomenological analysis of life in extra care housing

OBJECTIVES: To understand older adults' experiences of moving into extra care housing which offers enrichment activities alongside social and healthcare support. DESIGN: A longitudinal study was conducted which adopted a phenomenological approach to data generation and analysis. METHODS: Semi-structured interviews were conducted in the first 18 months of living in extra care housing. Interpretative phenomenological analysis was used because its commitment to idiography enabled an in-depth analysis of the subjective lived experience of moving into extra care housing. Themes generated inductively were examined against an existential-phenomenological theory of well-being. RESULTS: Learning to live in an extra care community showed negotiating new relationships was not straightforward; maintaining friendships outside the community became more difficult as capacity declined. In springboard for opportunity/confinement, living in extra care provided new opportunities for social engagement and a restored sense of self. Over time horizons began to shrink as incapacities grew. Seeking care illustrated reticence to seek care, due to embarrassment and a sense of duty to one's partner. Becoming aged presented an ontological challenge. Nevertheless, some showed a readiness for death, a sense of homecoming. CONCLUSIONS: An authentic later life was possible but residents required emotional and social support to live through the transition and challenges of becoming aged. Enhancement activities boosted residents' quality of life but the range of activities could be extended to cater better for quieter, smaller scale events within the community; volunteer activity facilitators could be used here. Peer mentoring may help build new relationships and opportunities for interactive stimulation. Acknowledging the importance of feeling-empathic imagination-in caregiving may help staff and residents relate better to each other, thus helping individuals to become ontologically secure and live well to the end.