Off-line signature verification and forgery detection system based on fuzzy modeling

This paper presents an innovative approach for signature verification and forgery detection based on fuzzy modeling. The signature image is binarized and resized to a fixed size window and is then thinned. The thinned image is then partitioned into a fixed number of eight sub-images called boxes. This partition is done using the horizontal density approximation approach. Each sub-image is then further resized and again partitioned into twelve further sub-images using the uniform partitioning approach. The features of consideration are normalized vector angle (α) from each box. Each feature extracted from sample signatures gives rise to a fuzzy set. Since the choice of a proper fuzzification function is crucial for verification, we have devised a new fuzzification function with structural parameters, which is able to adapt to the variations in fuzzy sets. This function is employed to develop a complete forgery detection and verification system.

Adaptive Signal Processing for Power Line Communications

This thesis represents a significant part of the research activity conducted during the PhD program in Information Technologies, supported by Selta S.p.A, Cadeo, Italy, focused on the analysis and design of a Power Line Communications (PLC) system. In recent times the PLC technologies have been considered for integration in Smart Grids architectures, as they are used to exploit the existing power line infrastructure for information transmission purposes on low, medium and high voltage lines. The characterization of a reliable PLC system is a current object of research as well as it is the design of modems for communications over the power lines. In this thesis, the focus is on the analysis of a full-duplex PLC modem for communication over high-voltage lines, and, in particular, on the design of the echo canceller device and innovative channel coding schemes.

Jobs, Unemployment, and Macroeconomic Transmission

Measuring Job Openings: Evidence from Swedish Plant Level Data. In modern macroeconomic models “job openings'' are a key component. Thus, when taking these models to the data we need an empirical counterpart to the theoretical concept of job openings. To achieve this, the literature relies on job vacancies measured either in survey or register data. Insofar as this concept captures the concept of job openings well we should see a tight relationship between vacancies and subsequent hires on the micro level. To investigate this, I analyze a new data set of Swedish hires and job vacancies on the plant level covering the period 2001-2012. I find that vacancies contain little power in predicting hires over and above (i) whether the number of vacancies is positive and (ii) plant size. Building on this, I propose an alternative measure of job openings in the economy. This measure (i) better predicts hiring at the plant level and (ii) provides a better fitting aggregate matching function vis-à-vis the traditional vacancy measure. Firm Level Evidence from Two Vacancy Measures. Using firm level survey and register data for both Sweden and Denmark we show systematic mis-measurement in both vacancy measures. While the register-based measure on the aggregate constitutes a quarter of the survey-based measure, the latter is not a super-set of the former. To obtain the full set of unique vacancies in these two databases, the number of survey vacancies should be multiplied by approximately 1.2. Importantly, this adjustment factor varies over time and across firm characteristics. Our findings have implications for both the search-matching literature and policy analysis based on vacancy measures: observed changes in vacancies can be an outcome of changes in mis-measurement, and are not necessarily changes in the actual number of vacancies. Swedish Unemployment Dynamics. We study the contribution of different labor market flows to business cycle variations in unemployment in the context of a dual labor market. To this end, we develop a decomposition method that allows for a distinction between permanent and temporary employment. We also allow for slow convergence to steady state which is characteristic of European labor markets. We apply the method to a new Swedish data set covering the period 1987-2012 and show that the relative contributions of inflows and outflows to/from unemployment are roughly 60/30. The remaining 10\% are due to flows not involving unemployment. Even though temporary contracts only cover 9-11\% of the working age population, variations in flows involving temporary contracts account for 44\% of the variation in unemployment. We also show that the importance of flows involving temporary contracts is likely to be understated if one does not account for non-steady state dynamics. The New Keynesian Transmission Mechanism: A Heterogeneous-Agent Perspective. We argue that a 2-agent version of the standard New Keynesian model---where a ``worker'' receives only labor income and a “capitalist'' only profit income---offers insights about how income inequality affects the monetary transmission mechanism. Under rigid prices, monetary policy affects the distribution of consumption, but it has no effect on output as workers choose not to change their hours worked in response to wage movements. In the corresponding representative-agent model, in contrast, hours do rise after a monetary policy loosening due to a wealth effect on labor supply: profits fall, thus reducing the representative worker's income. If wages are rigid too, however, the monetary transmission mechanism is active and resembles that in the corresponding representative-agent model. Here, workers are not on their labor supply curve and hence respond passively to demand, and profits are procyclical.

Adaptive back-propagation in on-line learning of multilayer networks

An adaptive back-propagation algorithm is studied and compared with gradient descent (standard back-propagation) for on-line learning in two-layer neural networks with an arbitrary number of hidden units. Within a statistical mechanics framework, both numerical studies and a rigorous analysis show that the adaptive back-propagation method results in faster training by breaking the symmetry between hidden units more efficiently and by providing faster convergence to optimal generalization than gradient descent.

Using ancillary statistics in on-line learning algorithms

Neural networks are usually curved statistical models. They do not have finite dimensional sufficient statistics, so on-line learning on the model itself inevitably loses information. In this paper we propose a new scheme for training curved models, inspired by the ideas of ancillary statistics and adaptive critics. At each point estimate an auxiliary flat model (exponential family) is built to locally accommodate both the usual statistic (tangent to the model) and an ancillary statistic (normal to the model). The auxiliary model plays a role in determining credit assignment analogous to that played by an adaptive critic in solving temporal problems. The method is illustrated with the Cauchy model and the algorithm is proved to be asymptotically efficient.

Globally optimal parameters for on-line learning in multilayer neural networks

We present a framework for calculating globally optimal parameters, within a given time frame, for on-line learning in multilayer neural networks. We demonstrate the capability of this method by computing optimal learning rates in typical learning scenarios. A similar treatment allows one to determine the relevance of related training algorithms based on modifications to the basic gradient descent rule as well as to compare different training methods.

Globally optimal on-line learning rules for multi-layer neural networks

We present a method for determining the globally optimal on-line learning rule for a soft committee machine under a statistical mechanics framework. This rule maximizes the total reduction in generalization error over the whole learning process. A simple example demonstrates that the locally optimal rule, which maximizes the rate of decrease in generalization error, may perform poorly in comparison.

On-line learning in radial basis functions networks

An analytic investigation of the average case learning and generalization properties of Radial Basis Function Networks (RBFs) is presented, utilising on-line gradient descent as the learning rule. The analytic method employed allows both the calculation of generalization error and the examination of the internal dynamics of the network. The generalization error and internal dynamics are then used to examine the role of the learning rate and the specialization of the hidden units, which gives insight into decreasing the time required for training. The realizable and over-realizable cases are studied in detail; the phase of learning in which the hidden units are unspecialized (symmetric phase) and the phase in which asymptotic convergence occurs are analyzed, and their typical properties found. Finally, simulations are performed which strongly confirm the analytic results.

Incorporating curvature information into on-line learning

We analyse the dynamics of a number of second order on-line learning algorithms training multi-layer neural networks, using the methods of statistical mechanics. We first consider on-line Newton's method, which is known to provide optimal asymptotic performance. We determine the asymptotic generalization error decay for a soft committee machine, which is shown to compare favourably with the result for standard gradient descent. Matrix momentum provides a practical approximation to this method by allowing an efficient inversion of the Hessian. We consider an idealized matrix momentum algorithm which requires access to the Hessian and find close correspondence with the dynamics of on-line Newton's method. In practice, the Hessian will not be known on-line and we therefore consider matrix momentum using a single example approximation to the Hessian. In this case good asymptotic performance may still be achieved, but the algorithm is now sensitive to parameter choice because of noise in the Hessian estimate. On-line Newton's method is not appropriate during the transient learning phase, since a suboptimal unstable fixed point of the gradient descent dynamics becomes stable for this algorithm. A principled alternative is to use Amari's natural gradient learning algorithm and we show how this method provides a significant reduction in learning time when compared to gradient descent, while retaining the asymptotic performance of on-line Newton's method.

Globally optimal on-line learning rules

We present a method for determining the globally optimal on-line learning rule for a soft committee machine under a statistical mechanics framework. This work complements previous results on locally optimal rules, where only the rate of change in generalization error was considered. We maximize the total reduction in generalization error over the whole learning process and show how the resulting rule can significantly outperform the locally optimal rule.

Sparse on-line Gaussian processes

We develop an approach for sparse representations of Gaussian Process (GP) models (which are Bayesian types of kernel machines) in order to overcome their limitations for large data sets. The method is based on a combination of a Bayesian online algorithm together with a sequential construction of a relevant subsample of the data which fully specifies the prediction of the GP model. By using an appealing parametrisation and projection techniques that use the RKHS norm, recursions for the effective parameters and a sparse Gaussian approximation of the posterior process are obtained. This allows both for a propagation of predictions as well as of Bayesian error measures. The significance and robustness of our approach is demonstrated on a variety of experiments.

Sparse on-line Gaussian processes

We develop an approach for sparse representations of Gaussian Process (GP) models (which are Bayesian types of kernel machines) in order to overcome their limitations for large data sets. The method is based on a combination of a Bayesian online algorithm together with a sequential construction of a relevant subsample of the data which fully specifies the prediction of the GP model. By using an appealing parametrisation and projection techniques that use the RKHS norm, recursions for the effective parameters and a sparse Gaussian approximation of the posterior process are obtained. This allows both for a propagation of predictions as well as of Bayesian error measures. The significance and robustness of our approach is demonstrated on a variety of experiments.

The development of a modified dissolution method suitable for investigating powder mixtures

A novel dissolution method was developed, suitable for powder mixtures, based on the USP basket apparatus. The baskets were modified such that the powder mixtures were retained within the baskets and not dispersed, a potential difficulty that may arise when using conventional USP basket and paddle apparatus. The advantages of this method were that the components of the mixtures were maintained in close proximity, maximizing any drug:excipient interaction and leading to more linear dissolution profiles. Two weakly acidic model drugs, ibuprofen and acetaminophen, and a selection of pharmaceutical excipients, including potential dissolution-enhancing alkalizing agents, were chosen for investigation. Dissolution profiles were obtained for simple physical mixtures. The f1 fit factor values, calculated using pure drug as the reference material, demonstrated a trend in line with expectations, with several dissolution enhancers apparent for both drugs. Also, the dissolution rates were linear over substantial parts of the profiles. For both drugs, a rank order comparison between the f1 fit factor and calculated dissolution rate, obtained from the linear section of the dissolution profile, demonstrated a correlation using a significance level of P=0.05. The method was proven to be suitable for discriminating between the effects of excipients on the dissolution of the model drugs. The method design produced dissolution profiles where the dissolution rate was linear for a substantial time, allowing determination of the dissolution rate without mathematical transformation of the data. This method may be suitable as a preliminary excipient-screening tool in the drug formulation development process.

Feasibility studies of an on-line identification technique

This thesis is concerned with the study of a non-sequential identification technique, so that it may be applied to the identification of process plant mathematical models from process measurements with the greatest degree of accuracy and reliability. In order to study the accuracy of the technique under differing conditions, simple mathematical models were set up on a parallel hybrid. computer and these models identified from input/output measurements by a small on-line digital computer. Initially, the simulated models were identified on-line. However, this method of operation was found not suitable for a thorough study of the technique due to equipment limitations. Further analysis was carried out in a large off-line computer using data generated by the small on-line computer. Hence identification was not strictly on-line. Results of the work have shovm that the identification technique may be successfully applied in practice. An optimum sampling period is suggested, together with noise level limitations for maximum accuracy. A description of a double-effect evaporator is included in this thesis. It is proposed that the next stage in the work will be the identification of a mathematical model of this evaporator using the teclmique described.

Q parameter estimation using numerical simulations for linear and nonlinear transmission systems

We compare the Q parameter obtained from scalar, semi-analytical and full vector models for realistic transmission systems. One set of systems is operated in the linear regime, while another is using solitons at high peak power. We report in detail on the different results obtained for the same system using different models. Polarisation mode dispersion is also taken into account and a novel method to average Q parameters over several independent simulation runs is described. © 2006 Elsevier B.V. All rights reserved.