Statistical mechanics of mutual information maximization

An unsupervised learning procedure based on maximizing the mutual information between the outputs of two networks receiving different but statistically dependent inputs is analyzed (Becker S. and Hinton G., Nature, 355 (1992) 161). By exploiting a formal analogy to supervised learning in parity machines, the theory of zero-temperature Gibbs learning for the unsupervised procedure is presented for the case that the networks are perceptrons and for the case of fully connected committees.

CDMA multiuser detection, neural networks, and statistical mechanics

A novel approach, based on statistical mechanics, to analyze typical performance of optimum code-division multiple-access (CDMA) multiuser detectors is reviewed. A `black-box' view ot the basic CDMA channel is introduced, based on which the CDMA multiuser detection problem is regarded as a `learning-from-examples' problem of the `binary linear perceptron' in the neural network literature. Adopting Bayes framework, analysis of the performance of the optimum CDMA multiuser detectors is reduced to evaluation of the average of the cumulant generating function of a relevant posterior distribution. The evaluation of the average cumulant generating function is done, based on formal analogy with a similar calculation appearing in the spin glass theory in statistical mechanics, by making use of the replica method, a method developed in the spin glass theory.

Developing a framework to evaluate business networks:the case of Ireland's industry-led network initiative

Business networks have been described as cooperative arrangements between independent business organisations that vary from contractual joint ventures to informal exchanges of information. This collaboration has become recognised as an innovative and efficient tool for organising interdependent activities, with benefits accruing to both firms and the local economy. For a number of years, resources have been devoted to supporting Irish networking policies. One recent example of such support is the Irish government's target of €20 million per annum for five years to support the creation of enterprise-led networks. It is imperative that a clear rationale for such interventions is established, as the opportunity cost of public funds is high. This article, therefore, develops an evaluation framework for such networking interventions. This framework will facilitate effective programme planning, implementation and evaluation. It will potentially show how a chain of cause-and-effect at both micro and macro-levels for networking interventions can be established.

To reveal is to critique:actor-network theory and critical information systems research

It has been suggested that, in order to maintain its relevance, critical research must develop a strong emphasis on empirical work rather than the conceptual emphasis that has typically characterized critical scholarship in management. A critical project of this nature is applicable in the information systems (IS) arena, which has a growing tradition of qualitative inquiry. Despite its relativist ontology, actor–network theory places a strong emphasis on empirical inquiry and this paper argues that actor–network theory, with its careful tracing and recording of heterogeneous networks, is well suited to the generation of detailed and contextual empirical knowledge about IS. The intention in this paper is to explore the relevance of IS research informed by actor–network theory in the pursuit of a broader critical research project as de? ned in earlier work.

Market-level information and the diffusion of competing technologies:an exploratory analysis of the LAN industry

Market-level information diffused by print media may contribute to the legitimation of an emerging technology and thus influence the diffusion of competing technological standards. After analyzing more than 10,000 trade media abstracts from the Local Area Networks (LAN) industry published between 1981 and 2000, we found the presence of differential effects on the adoption of competing standards by two market-level information types: technology and product availability. The significance of these effects depends on the technology's order of entry and suggests that high-tech product managers should make strategic use of market-level information by appropriately focusing the content of their communications. © 2007 Elsevier B.V. All rights reserved.

The relationship between market share and information in a high-tech industry

The role of information in high-technology markets is critical (Dutta, Narasimhan and Rajiv 1999; Farrell and Saloner 1986; Weiss and Heide 1993). In these markets, the volatility and volume of information present managers and researchers with the considerable challenge of monitoring such information and examining how potential customers may respond to it. This article examines the effects of the type and volume of information on the market share of different technological standards in the Local Area Networks (LAN) industry. We identify three different types of information: technological, availability and adoption. Our empirical application suggests that all three types of information have significant effects on the market share of a technological standard, but their direction and magnitude differ. More specifically, technology-related information is negatively related to market share as it demonstrates that the underlying technology is immature and still evolving. Both availability and adoption-related information have a positive effect on market share, but the former is larger than the latter. We conclude that high-tech firms should emphasize the dissemination of information, especially availability-related, as part of their promotional strategy for a new technology. Otherwise, they may risk missing an opportunity to achieve a higher share and establish their market presence.

Neural networks: A principled perspective

Introductory accounts of artificial neural networks often rely for motivation on analogies with models of information processing in biological networks. One limitation of such an approach is that it offers little guidance on how to find optimal algorithms, or how to verify the correct performance of neural network systems. A central goal of this paper is to draw attention to a quite different viewpoint in which neural networks are seen as algorithms for statistical pattern recognition based on a principled, i.e. theoretically well-founded, framework. We illustrate the concept of a principled viewpoint by considering a specific issue concerned with the interpretation of the outputs of a trained network. Finally, we discuss the relevance of such an approach to the issue of the validation and verification of neural network systems.

Cultural impact of wireless networks:emerging experiences in the healthcare sector

Culture defines collective behavior and interactions among people in groups. In organizations, it shapes group identity, work pattern, communication schemes, and interpersonal relations. Any change in organizational culture will lead to changes in these elements of organizational factors, and vice versa. From a managerial standpoint, how to cultivate an organizational culture that would enhance these aforementioned elements in organizational workplace should thus be taken into serious consideration. Based on cases studies in two hospitals, this paper investigates how organizational culture is shaped by a particular type of information and communication technology, wireless networks, a topic that is generally overlooked by the mainstream research community, and in turn implicates how such cultural changes in organizations renovate their competitiveness in the marketplace. Lessons learned from these cases provide valuable insights to emerging IT management and culture studies in general and in wireless network management in the healthcare sector in particular.

Packet voice/data in local broadcast networks

A local area network that can support both voice and data packets offers economic advantages due to the use of only a single network for both types of traffic, greater flexibility to changing user demands, and it also enables efficient use to be made of the transmission capacity. The latter aspect is very important in local broadcast networks where the capacity is a scarce resource, for example mobile radio. This research has examined two types of local broadcast network, these being the Ethernet-type bus local area network and a mobile radio network with a central base station. With such contention networks, medium access control (MAC) protocols are required to gain access to the channel. MAC protocols must provide efficient scheduling on the channel between the distributed population of stations who want to transmit. No access scheme can exceed the performance of a single server queue, due to the spatial distribution of the stations. Stations cannot in general form a queue without using part of the channel capacity to exchange protocol information. In this research, several medium access protocols have been examined and developed in order to increase the channel throughput compared to existing protocols. However, the established performance measures of average packet time delay and throughput cannot adequately characterise protocol performance for packet voice. Rather, the percentage of bits delivered within a given time bound becomes the relevant performance measure. Performance evaluation of the protocols has been examined using discrete event simulation and in some cases also by mathematical modelling. All the protocols use either implicit or explicit reservation schemes, with their efficiency dependent on the fact that many voice packets are generated periodically within a talkspurt. Two of the protocols are based on the existing 'Reservation Virtual Time CSMA/CD' protocol, which forms a distributed queue through implicit reservations. This protocol has been improved firstly by utilising two channels, a packet transmission channel and a packet contention channel. Packet contention is then performed in parallel with a packet transmission to increase throughput. The second protocol uses variable length packets to reduce the contention time between transmissions on a single channel. A third protocol developed, is based on contention for explicit reservations. Once a station has achieved a reservation, it maintains this effective queue position for the remainder of the talkspurt and transmits after it has sensed the transmission from the preceeding station within the queue. In the mobile radio environment, adaptions to the protocols were necessary in order that their operation was robust to signal fading. This was achieved through centralised control at a base station, unlike the local area network versions where the control was distributed at the stations. The results show an improvement in throughput compared to some previous protocols. Further work includes subjective testing to validate the protocols' effectiveness.

Transmission of image data over digital networks involving mobile terminals

There is a growing demand for data transmission over digital networks involving mobile terminals. An important class of data required for transmission over mobile terminals is image information such as street maps, floor plans and identikit images. This sort of transmission is of particular interest to the service industries such as the Police force, Fire brigade, medical services and other services. These services cannot be applied directly to mobile terminals because of the limited capacity of the mobile channels and the transmission errors caused by the multipath (Rayleigh) fading. In this research, transmission of line diagram images such as floor plans and street maps, over digital networks involving mobile terminals at transmission rates of 2400 bits/s and 4800 bits/s have been studied. A low bit-rate source encoding technique using geometric codes is found to be suitable to represent line diagram images. In geometric encoding, the amount of data required to represent or store the line diagram images is proportional to the image detail. Thus a simple line diagram image would require a small amount of data. To study the effect of transmission errors due to mobile channels on the transmitted images, error sources (error files), which represent mobile channels under different conditions, have been produced using channel modelling techniques. Satisfactory models of the mobile channel have been obtained when compared to the field test measurements. Subjective performance tests have been carried out to evaluate the quality and usefulness of the received line diagram images under various mobile channel conditions. The effect of mobile transmission errors on the quality of the received images has been determined. To improve the quality of the received images under various mobile channel conditions, forward error correcting codes (FEC) with interleaving and automatic repeat request (ARQ) schemes have been proposed. The performance of the error control codes have been evaluated under various mobile channel conditions. It has been shown that a FEC code with interleaving can be used effectively to improve the quality of the received images under normal and severe mobile channel conditions. Under normal channel conditions, similar results have been obtained when using ARQ schemes. However, under severe mobile channel conditions, the FEC code with interleaving shows better performance.

Novel hydrogel copolymers and semi-interpenetraing polymer networks

Interpenetrating polymer networks (lPN's), have been defined as a combination of two polymers each in network form, at least one of which has been synthesised and / or crosslinked in the presence of the other. A semi-lPN, is formed when only one of the polymers in the system is crosslinked, the other being linear. lPN's have potential advantages over homogeneous materials presently used in biomedical applications, in that their composite nature gives them a useful combination of properties. Such materials have potential uses in the biomedical field, specifically for use in hard tissue replacements, rigid gas permeable contact lenses and dental materials. Work on simply two or three component systems in both low water containing lPN's supplemented by the study of hydrogels (water swollen hydrophilic polymers) can provide information useful in the future development of more complex systems. A range of copolymers have been synthesised using a variety of methacrylates and acrylates. Hydrogels were obtained by the addition of N-vinyl pyrrolidone to these copolymers. A selection of interpenetrants were incorporated into the samples and their effect on the copolymer properties was investigated. By studying glass transition temperatures, mechanical, surface, water binding and oxygen permeability properties samples were assessed for their suitability for use as biomaterials. In addition copolymers containing tris-(trimethylsiloxy)-y-methacryloxypropyl silane, commonly abbreviated to 'TRlS', have been investigated. This material has been shown to enhance oxygen permeability, a desirable property when considering the design of contact lenses. However, 'TRIS' has a low polar component of surface free energy and hence low wettability. Copolymerisation with a range of methacrylates has shown that significant increases in surface wettability can be obtained without a detrimental effect on oxygen permeability. To further enhance to surface wettability 4-methacryloxyethyl trimellitic anhydride was incorporated into a range of promising samples. This study has shown that by careful choice of monomers it is possible to synthesise polymers that possess a range of properties desirable in biomedical applications.

A novel silicon membrane-based biosensing platform using distributive sensing strategy and artificial neural networks for feature analysis

A novel biosensing system based on a micromachined rectangular silicon membrane is proposed and investigated in this paper. A distributive sensing scheme is designed to monitor the dynamics of the sensing structure. An artificial neural network is used to process the measured data and to identify cell presence and density. Without specifying any particular bio-application, the investigation is mainly concentrated on the performance testing of this kind of biosensor as a general biosensing platform. The biosensing experiments on the microfabricated membranes involve seeding different cell densities onto the sensing surface of membrane, and measuring the corresponding dynamics information of each tested silicon membrane in the form of a series of frequency response functions (FRFs). All of those experiments are carried out in cell culture medium to simulate a practical working environment. The EA.hy 926 endothelial cell lines are chosen in this paper for the bio-experiments. The EA.hy 926 endothelial cell lines represent a particular class of biological particles that have irregular shapes, non-uniform density and uncertain growth behaviour, which are difficult to monitor using the traditional biosensors. The final predicted results reveal that the methodology of a neural-network based algorithm to perform the feature identification of cells from distributive sensory measurement has great potential in biosensing applications.

Studies on the generalisation of Gaussian processes and Bayesian neural networks

The assessment of the reliability of systems which learn from data is a key issue to investigate thoroughly before the actual application of information processing techniques to real-world problems. Over the recent years Gaussian processes and Bayesian neural networks have come to the fore and in this thesis their generalisation capabilities are analysed from theoretical and empirical perspectives. Upper and lower bounds on the learning curve of Gaussian processes are investigated in order to estimate the amount of data required to guarantee a certain level of generalisation performance. In this thesis we analyse the effects on the bounds and the learning curve induced by the smoothness of stochastic processes described by four different covariance functions. We also explain the early, linearly-decreasing behaviour of the curves and we investigate the asymptotic behaviour of the upper bounds. The effect of the noise and the characteristic lengthscale of the stochastic process on the tightness of the bounds are also discussed. The analysis is supported by several numerical simulations. The generalisation error of a Gaussian process is affected by the dimension of the input vector and may be decreased by input-variable reduction techniques. In conventional approaches to Gaussian process regression, the positive definite matrix estimating the distance between input points is often taken diagonal. In this thesis we show that a general distance matrix is able to estimate the effective dimensionality of the regression problem as well as to discover the linear transformation from the manifest variables to the hidden-feature space, with a significant reduction of the input dimension. Numerical simulations confirm the significant superiority of the general distance matrix with respect to the diagonal one.In the thesis we also present an empirical investigation of the generalisation errors of neural networks trained by two Bayesian algorithms, the Markov Chain Monte Carlo method and the evidence framework; the neural networks have been trained on the task of labelling segmented outdoor images.

The application of artificial neural networks to the interpretation and classification of freshwater benthic invertebrate communities

This thesis presents a thorough and principled investigation into the application of artificial neural networks to the biological monitoring of freshwater. It contains original ideas on the classification and interpretation of benthic macroinvertebrates, and aims to demonstrate their superiority over the biotic systems currently used in the UK to report river water quality. The conceptual basis of a new biological classification system is described, and a full review and analysis of a number of river data sets is presented. The biological classification is compared to the common biotic systems using data from the Upper Trent catchment. This data contained 292 expertly classified invertebrate samples identified to mixed taxonomic levels. The neural network experimental work concentrates on the classification of the invertebrate samples into biological class, where only a subset of the sample is used to form the classification. Other experimentation is conducted into the identification of novel input samples, the classification of samples from different biotopes and the use of prior information in the neural network models. The biological classification is shown to provide an intuitive interpretation of a graphical representation, generated without reference to the class labels, of the Upper Trent data. The selection of key indicator taxa is considered using three different approaches; one novel, one from information theory and one from classical statistical methods. Good indicators of quality class based on these analyses are found to be in good agreement with those chosen by a domain expert. The change in information associated with different levels of identification and enumeration of taxa is quantified. The feasibility of using neural network classifiers and predictors to develop numeric criteria for the biological assessment of sediment contamination in the Great Lakes is also investigated.

Drying processess in the United Kingdom:assessment of industrial energy utilisation and efficiency of drying systems and the modelling of drying characters using neural networks (BL)

Drying is an important unit operation in process industry. Results have suggested that the energy used for drying has increased from 12% in 1978 to 18% of the total energy used in 1990. A literature survey of previous studies regarding overall drying energy consumption has demonstrated that there is little continuity of methods and energy trends could not be established. In the ceramics, timber and paper industrial sectors specific energy consumption and energy trends have been investigated by auditing drying equipment. Ceramic products examined have included tableware, tiles, sanitaryware, electrical ceramics, plasterboard, refractories, bricks and abrasives. Data from industry has shown that drying energy has not varied significantly in the ceramics sector over the last decade, representing about 31% of the total energy consumed. Information from the timber industry has established that radical changes have occurred over the last 20 years, both in terms of equipment and energy utilisation. The energy efficiency of hardwood drying has improved by 15% since the 1970s, although no significant savings have been realised for softwood. A survey estimating the energy efficiency and operating characteristics of 192 paper dryer sections has been conducted. Drying energy was found to increase to nearly 60% of the total energy used in the early 1980s, but has fallen over the last decade, representing 23% of the total in 1993. These results have demonstrated that effective energy saving measures, such as improved pressing and heat recovery, have been successfully implemented since the 1970s. Artificial neural networks have successfully been applied to model process characteristics of microwave and convective drying of paper coated gypsum cove. Parameters modelled have included product moisture loss, core gypsum temperature and quality factors relating to paper burning and bubbling defects. Evaluation of thermal and dielectric properties have highlighted gypsum's heat sensitive characteristics in convective and electromagnetic regimes. Modelling experimental data has shown that the networks were capable of simulating drying process characteristics to a high degree of accuracy. Product weight and temperature were predicted to within 0.5% and 5C of the target data respectively. Furthermore, it was demonstrated that the underlying properties of the data could be predicted through a high level of input noise.