This work presents the design of a mobile gateway for independent life and e-health support. Technologies such as OSGi and Web Services have been used to deploy several services as an example, such as bio-medical parameter monitorization, alerts, and communication with a coordination center. A complete functional service which uses context-awareness has been explained and the benefits of using Service Oriented Architecture are explained. Finally, the lessons learned of this kind of development, which involves several types of systems, data, languages, and development groups, are discussed.

Multiple sequence alignment (MSA) is one of the most studied approach in Bioinformatics to carry out other outstanding tasks like structural predictions, biological function analysis or next-generation sequencing. However, MSA algorithms do not achieve consistent results in all cases, as alignments become difficult when sequences have low similarity. In other words, each algorithm is focused in specific features of sequences and their results depend on them. For this reason, each approach could align better those sections of sequences that include such features, obtaining partially optimal solutions. In this work, a multiobjective evolutionary algorithm based on NSGA-II will be implemented in order to assemble previously aligned sequences, trying to avoid suboptimal alignments.

In this paper, a new approach for Self-evolving PArameter-free fuzzy Rule-based Controller (SPARC) is proposed. Two illustrative examples are provided aiming a proof of concept. The proposed controller can start with no pre-defined fuzzy rules, and does not need to pre-define the range of the output or control variables. This SPARC learns autonomously from its own actions while performing the control of the plant. It does not use any parameters, explicit membership functions, any off-line pre-training nor the explicit model (e.g. in a form of differential equations) of the plant. It combines the relative older concept of indirect adaptive control with the newer concepts of (self-)evolving fuzzy rule-based systems (and controllers, in particular) and with the very recent concept of parameter-free, data cloud and data density based fuzzy rule based systems (and controllers in particular). It has been demonstrated that a fully autonomously and in an unsupervised manner (based only on the data density and selecting representative prototypes/focal points from the control hyper-surface acting as a data space) it is possible generate a parameter-free control structure and evolve it in on-line mode. Moreover, the results demonstrate that this autonomous controller is effective (has comparative error and performance characteristics) to other known controllers, including self-learning ones, but surpasses them with its flexibility and extremely lean structure (small number of prototypes/focal points which serve as seeds to form parameter-free and membership function-free fuzzy rules based on them). The illustrative examples aim primarily proof of concept.

There are several dementias but Alzheimer's Disease (AD) is leading cause of dementia in the world. In this paper, a new methodology for classification of MR images is proposed, using a large data base (more than one thousand patient). We have two main objectives in this paper: a first one where a classification method is developed to classify MR images as either normal or with the Alzheimer's disease and a second which is the ambitious goal of this study, the identification and classification between normal subjects, MCI patients and AD patients. It is noteworthy that with this last study we could offer a tool to assist the early diagnosis of dementia.

Fully auditory Brain-computer interfaces based on the dichotic listening task (DL-BCIs) are suited for users unable to do any muscular movement, which includes gazing, exploration or coordination of their eyes looking for inputs in form of feedback, stimulation or visual support. However, one of their disadvantages, in contrast with the visual BCIs, is their lower performance that makes them not adequate in applications that require a high accuracy. To overcome this disadvantage, we employed a Bayesian approach in which the DL-BCI was modeled as a Binary phase shift keying receiver for which the accuracy can be estimated a priori as a function of the signal-to-noise ratio. The results showed the measured accuracy to match the predefined target accuracy, thus validating this model that made possible to estimate in advance the classification accuracy on a trial-by-trial basis. This constitutes a novel methodology in the design of fully auditory DL-BCIs that let us first, define the target accuracy for a specific application and second, classify when the signal-to-noise ratio guarantees that target accuracy.

Present trend towards multiprocessor and multi-core architectures as well as programmable NICs (Network Interface Cards) provides new opportunities to exploit the available parallelism in the network interface design and implementation to cope with the high communication overhead required to take advantage of a multi-gigabit links. In this paper we have used dynamic web servers to evaluate several alternatives to distribute the network interface work among the different cores available in the node. These alternatives have been devised according to the affinity of the work to be done with the location (proximity to the memories where the different data structures are stored) and characteristics of the processing core. Improvements in the throughput (up to 110%), the response time (up to 500%) and the requests attended per second (up to 85%) have been observed in our experiments on dynamic web servers.

This paper proposes using the ANOVA (ANalysis Of the VAriance) method to carry out an exhaustive analysis of the simulated annealing (Sim-Ann) method and the different parameters it requires, such as those related to: the neighbourhood; the cooling scheme; the initial temperature; the number of times the cooling scheme is applied; and the number of times we search for best individual before the temperature is cooled. When undertaking a detailed statistical analysis of the influence of each parameter, the designer should pay attention mostly to the parameter presenting values that are statistically most significant. Following this idea, the significance and relative importance of the parameters with respect to the obtained results, as well as suitable values for each of these, were obtained using ANOVA on four well known function optimization problems.

Brain-computer interfaces based on steady-state visual evoked potential (SSVEP-BCIs) achieve a high performance due to their multiclass nature in paradigms where gazing is needed. Studies of binary SSVEP-BCIs have been presented without the need of gazing at the expense of low performance. This study presents a high performance binary SSVEP-BCI that allows an efficient communication without the need of gazing. The information transmission rate (ITR) obtained was 0.64±0.27 bits per seconds with peaks of 1.16 bits per second and an accuracy of 90±7%, which is an excellent performance for a binary SSVEP-BCI. To achieve this performance two main factors were involved: on the one hand, the use of both amplitude and phase of the SSVEP in the classification and, on the other, the use of the absence of gaze (“thousand-yard stare”) as a way of helping to ignore the stimulus.

This paper describes an artificial neural network based system for classifying the contents of hyperspectral images that is able to automatically reduce the dimensionality of the data provided by the hyperspectrometers without compromising their efficacy. The data reduction is achieved through the adaptation of the window size and the number of parameters that make up the description of the spectral signatures within the window as training progresses. Following this approach, a user just needs to specify the minimum resolution desired on the output or category image and the level of discrimination among categories, and the system will try to meet these requirements by modifying during training the size and number of inputs to the network. When it is not possible to comply with both requirements, the system will provide a compromise solution that minimizes the global discrimination error, which takes into account the spatial discrimination and the discrimination among classes.

The availability of multi-core processors and programmable NICs (Network Interface Cards), such as TOEs (TCP/IP Offloading Engines), provides new opportunities for designing efficient network interfaces to cope with the gap between the improvement rates of link bandwidths and microprocessor performance. This gap poses important challenges related with the high computational requirements associated to the traffic volumes and wider functionality that the network interface has to support. An opportunity to reach these goals comes from the exploitation of the parallelism in the communication path by distributing the protocol processing work across processors available in the computer, i.e. multi-core microprocessors and programmable NICs. Thus, alternatives such as offloading and onloading try to release host CPU cycles by this approach. Nevertheless, whereas onloading uses another general-purpose processor, either included in a multi-core microprocessor or in a symmetric multiprocessor (SMP), offloading takes advantage of processors in programmable network interface cards (NICs). Some experimental results demonstrate that the relative improvement on peak throughput offered by offloading and onloading depends on the rate of application workload to communication overhead, the message sizes, and on the characteristics of system architecture, more specifically the bandwidth of the buses and the way the NIC is connected to the system processor and memory. Thus, in this paper we propose a network interface that takes the advantages of both offloading and onloading approaches while avoids their respective drawbacks. The performance analyses done by using a full-system simulator, shows that, in the benchmarks and application used for the experiments, our hybrid interface improves the latency and bandwidth behavior of the onloading and offloading approaches.

In many real world applications type I (false positive) and type II (false negative) errors have to be dealt with separately, which is a complex problem since an attempt to minimize one of them usually makes the other grow. In fact, a type of error can be more important than the other, and a trade-off that minimizes the most important error type must be reached. In the case of the bankruptcy prediction problem the error type II is of greater importance, being unable to identify that a company is at risk causes problems to creditors and slows down the taking of measures that may solve the problem. Despite the importance of type II errors, most bankruptcy prediction methods take into account only the global classification error. In this paper we propose and compare two methods to optimize both error types in classification: artificial neural networks and function trees ensembles created through multiobjective optimization. Since the multiobjective optimization process produces a set of equally optimal results (Pareto front) the classification of the test patterns in both cases is based on the non-dominated solutions acting as an ensemble. The experiments prove that, although the best classification rates are obtained using the artificial neural network, the multiobjective genetic programming model is able to generate comparable results in the form of an analytical function.

In a connected world, spare CPU cycles are up for grabs, if you only make its obtention easy enough. In this paper we present a distributed evolutionary computation system that uses the computational capabilities of the ubiquituous web browser. Asynchronous Javascript and JSON (Javascript Object Notation, a serialization protocol) allows anybody with a web browser (that is, mostly everybody connected to the Internet) to participate in a genetic algorithm experiment with little effort, or none at all. Since, in this case, computing becomes a social activity and is inherently impredictable, in this paper we will explore the performance of this kind of virtual computer by solving simple problems such as the Royal Road function and analyzing how many machines and evaluations it yields. We will also examine possible performance bottlenecks and how to solve them, and, finally, issue some advice on how to set up this kind of experiments to maximize turnout and, thus, performance. The experiments show that we we can obtain high, and to a certain point, reliable performance from volunteer computing based on AJAJ, with speedups of up to several (averaged) machines.

The realization of a control unit can be done using a complex circuitry or microprogramming. The latter may be considered as an alternative method of implementation of machine instructions that can reduce the complexity and increase the flexibility of the control unit. The microcode efficiency and speed are of vital importance for the computer to execute machine instructions fast. This is a difficult task and it requires expert knowledge. It would be interesting and very helpful to have automated tools that, given a machine instruction description, could generate an efficient and correct microprogram. A good option is to use evolutionary computation techniques, which have proved been effective in the evolution of computer programs. In this paper, we intend to show how evolutionary computing techniques could be used to face this problem of generating efficient microprograms. We have developed a microarchitecture simulator of a real machine in order to evaluate an individual and to assign it the fitness value (to determine whether this candidate solution correctly implements the instruction machine). The proposed method is successful in generating correct solutions, not only for the machine code instruction set, but for new machine instructions not included in such set. We have shown that our approach can generate microprogramms to execute (to schedule microinstructions) the machine level instructions for a real machine. Moreover, this evolutive method could be applied to any microarchitecture just by changing the microinstruction set and pre-conditions of each machine instruction to guide evolution.

Selective attention to visual or auditory stimuli that elicits steady-state visual or auditory responses (SSVEP or ASSR respectively) amplifies the power of those flickering frequencies of the stimuli measured in the electro-encephalography (EEG). The design of brain-computer interfaces (BCI) based on selective attention to auditory stimuli that elicits ASSRs has two major advantages: First, no much training is needed as selective attention is an inherent skill of human beings. Second, as the human hearing system behaves as an amplitude modulated (AM) envelope detector, if the auditory stimuli are previously AM modulated, a steady-state response signal can be easily elicited. As EEG signals are considered to be non stationary, this is an interesting approach to EEG analysis as the feature of our interest can be considered stationary, hence Fourier analysis applied. In this paper we study the behavior of a Perceptron as a classifier of a BCI system where the features for classification are extracted from EEG as responses to AM modulated auditory stimuli and how we can use it in a BCI system.

This paper compares the onloading and offloading alternatives for improving up communication. Both strategies try to release host CPU cycles by taking advantage of the execution of the communication workload in other processors present in the node. Nevertheless, whereas onloading uses another general-purpose processor, either included in a chip multiprocessor (CMP) or in a symmetric multiprocessor (SMP), offloading takes advantage of processors in programmable network interface cards (NICs). Here, it is shown that the relative improvement on peak throughput offered by offloading and onloading depends on the rate of application workload to communication overhead, the message sizes, and the characteristics of system architecture, more specifically the buses bandwidth and the way the NIC is connected to the system processor and memory. In our implementations, offloading provides lower latencies than onloading although the CPU utilization and interrupts are lower for onloading.

Due to the network technology advances, an order-of-magnitude jump has been produced in the network bandwidth. This fact has reawake the interest in protocol offloading (particularly in protocols such as TCP, that require a lot of CPU resources to process the stack), since network communication is one key factor for the system performance. Nevertheless, there are controversial studies on protocol offloading benefits, and much research work on offloading, particularly on TCP offloading as it has been the main transport protocol for many years, and in fact, it is the most used protocol in the networks. This paper describes a model to simulate by Simics the protocol offloading technique to keep low the communication overheads on the host CPU. The simulation platform provided by Simics can be used for a functional system simulation, including the application program, the operating system, the protocol stack and the device drivers, since network-oriented applications require a full-system simulation. As Simics does not provide a detailed network I/O model, in this paper we describe the way we have generated a model that allows the simulation of the offloading effects

We present the hardware implementation of a simple, fast technique for depth estimation based on phase measurement. This technique avoids the problem of phase warping and is much less susceptible to camera noise and distortion than standard block-matching stereo systems. The architecture exploits the parallel computing resources of FPGA devices to achieve a computation speed of 65 megapixels per second. For this purpose, we have designed a fine-grain pipeline structure that can be arranged with a customized frame-grabber module to process 52 frames per second at a resolution of 1280 x 960 pixels. We have measured the system's degradation due to bit quantization errors and compared its performance with other previous approaches. We have also used different Gabor-scale circuits, which can be selected by the user according to the application addressed and typical image structure in the target scenario.

This paper describes our work with Simics (Magnusson, 2002) in the evaluation of protocol offloading to keep low the communication overheads on the host CPU. Simics is a platform that can be used for a functional system simulation, including the application program, the operating system, the protocol stack and the device drivers. Nevertheless, network-oriented applications require a full-system simulation that includes not only the OS code, but also a model of the memory system and a detailed timing model of network DMA effects (Binkert, 2003). As Simics does not provide a detailed network I/O model, it could present some limitations that make it difficult an accurate network-oriented full-system simulation. Here we present the way we have overcome this problem

Radial basis function neural networks (RBFNNs) have been applied to solve problems of classification, function approximation and time series prediction. In the design of an RBFNN it is necessary to set the values for the positions of the centers and the radii for each RBF. In the literature it is usually performed an initialization step to set the positions of the centers and, once they are placed, the radii are calculated using a heuristic. In this paper, a new algorithm to set the value of those two parameters is presented. This new algorithm uses a supervised learning in such a way that the position of the centers will be constrained by the output of the function to be approximated. Since each center represents a neuron that is activated by the input vectors, the radii are initialized using the center's positions and their activation grades. In this way, the calculation of the radii is also influenced by the output of the target function, not like in other heuristics where only the positions of the centers or the input vectors are considered. As the experiments show, the new algorithm outperforms other algorithms previously used for this problem.

The logarithmic image processing model (LIP) is a robust mathematical framework, which, among other benefits, behaves invariantly to illumination changes. This paper presents, for the first time, two general formulations of the 2-D convolution of separable kernels under the LIP paradigm. Although both formulations are mathematically equivalent, one of them has been designed avoiding the operations which are computationally expensive in current computers. Therefore, this fast LIP convolution method allows to obtain significant speedups and is more adequate for real-time processing. In order to support these statements, some experimental results are shown in Section V.

The goal of this paper is to achieve real time control of a monotonic system which, in general, may be non-linear and whose differential equations are unknown. We assume that there is no model of the plant available so there cannot be any off-line pre-training of the main controller parameters. We propose a both adaptive and self-learning algorithm capable of starting from a "void" fuzzy controller and, in real time, optimizing the fuzzy controller's rules (both antecedents and consequents) in order to translate the state of the plant to the desired value in the shortest possible time.

A novel approach to achieve real-time global learning in fuzzy controllers is proposed. Both the rule consequents and the membership functions defined in the premises of the fuzzy rules are tuned using a one-step algorithm, which is capable of controlling nonlinear plants with no prior offline training. Direct control is achieved by means of two auxiliary systems: The first one is responsible for adapting the consequents of the main controller's rules to minimize the error arising at the plant output, while the second auxiliary system compiles real input-output data obtained from the plant. The system then learns in real time from these data taking into account, not the current state of the plant but rather the global identification performed. Simulation results show that this approach leads to an enhanced control policy thanks to the global learning performed, avoiding overfitting.

The identification of a model is one of the key issues in the field of fuzzy system modeling and function approximation theory. There are numerous approaches to the issue of parameter optimization within a fixed fuzzy system structure but no reliable method to obtain the optimal topology of the fuzzy system from a set of input-output data. This paper presents a reliable method to obtain the structure of a complete rule-based fuzzy system for a specific approximation accuracy of the training data, i.e., it can decide which input variables must be taken into account in the fuzzy system and how many membership functions (MFs) are needed in every selected input variable in order to reach the approximation target with the minimum number of parameters.

To date, clustering techniques have always been oriented to solve classification and pattern recognition problems. However, some authors have applied them unchanged to construct initial models for function approximators. Nevertheless, classification and function approximation problems present quite different objectives. Therefore it is necessary to design new clustering algorithms specialized in the problem of function approximation. This paper presents a new clustering technique, specially designed for function. approximation problems, which improves the performance of the approximator system obtained, compared with other models derived from traditional classification oriented clustering algorithms and input-output clustering techniques.

Interest in hybrid methods that combine artificial neural networks and evolutionary algorithms has grown in the last few years, due to their robustness and ability to design networks by setting initial weight values, by searching the architecture and the learning rule and parameters. This paper presents an exhaustive analysis of the G-Prop method, and the different parameters the method requires (population size, selection rate, initial weight range, number of training epochs, etc.) are determined. The paper also the discusses the influence of the application of genetic operators on the precision (classification ability or error) and network size in classification problems. The significance and relative importance of the parameters with respect to the results obtained, as well as suitable values for each, were obtained using the ANOVA (analysis of the variance). Experiments show the significance of parameters concerning the neural network and learning in the hybrid methods. The parameters found using this method were used to compare the G-Prop method both to itself with other parameter settings, and to other published methods.

The CLIC is a lightweight protocol that has recently been proposed for efficient communication in clusters using the Linux Operating System. Besides optimizing communication performance, by reducing the latencies and increasing the bandwidth figures even for short messages, the proposed communication layer also meets other requirements such as multiprogramming, portability, protection against corrupted programs, reliable message delivery, direct access to the network for all applications, etc. In this way, instead of removing the operating system kernel from the critical path and creating a user-level network interface, the operating system support has been optimized to provide reliable and efficient network software, avoiding the TCP/IP protocol stack. The LAM-MPI communication layer has been implemented on top of the proposed protocol and the communication performance has been tested in a cluster of PCs with Linux OS and interconnected with Fast Ethernet.

An enhanced self-organizing controller (E-SOC) is presented to achieve real time global learning in fuzzy controllers. Direct control is achieved by means of two auxiliary systems: the first one is responsible for adapting the consequents of the main controller's rules to minimize the error arising at the plant output, while the second auxiliary system compiles real input/output data obtained from the plant. The system then learns in real time from these data taking into account, not the current state of the plant but rather the global identification performed. Simulation results show that this approach leads to an enhanced control policy thanks to the global learning performed.

A new approach is presented to fine tune a fuzzy controller based on very limited information on the nonlinear plant to be controlled. Without any off-line pre-training, the algorithm achieves very high control performance through a two-stage algorithm. In the first stage, coarse tuning of the fuzzy rules (both rule consequents and membership functions of the premises) is accomplished in real time using a small amount of information about the plant and an overall analysis of the main operating regions. In stage two, fine tuning of the fuzzy rules is achieved based on the controller output error using a gradient-based method.

This paper presents a reliable method to obtain the structure of a complete rule-based fuzzy system for a specific approximation accuracy of the training data, i.e. it can decide which input variables must be taken into account in the fuzzy system and how many membership functions are needed in every selected input variable in order to reach the approximation target with the minimum number of parameters.

An analysis of the influence of weight and input perturbations in a multilayer perceptron (MLP) is made in this article. Quantitative measurements of fault tolerance, noise immunity, and generalization ability are provided. From the expressions obtained, it is possible to justify some previously reported conjectures and experimentally obtained results (e.g., the influence of weight magnitudes, the relation between training with noise and the generalization ability, the relation between fault tolerance and the generalization ability). The measurements introduced here are explicitly related to the mean squared error degradation in the presence of perturbations, thus constituting a selection criterion between different alternatives of weight configurations. Moreover, they allow us to predict the degradation of the learning performance of an MLP when its weights or inputs are deviated from their nominal values and thus, the behavior of a physical implementation can be evaluated before the weights are mapped on it according to its accuracy.

Genetic algorithms, time series prediction, and Monte Carlo simulation are applied to dynamic programming in order to solve complex planning and control problems in which decisions are made in stages, and the states and control belong to a continuous space. Each decision has an immediate associated cost and also affects the cost of future stages. Therefore, a balance is required between a low cost solution at the present and the possible high costs in the future. A hybrid genetic algorithm is used to determine the feasible functioning states in each stage. A procedure for series prediction based on RBF networks allows the uncertainty about state transitions to be avoided and Monte Carlo simulations are used to approximate the cost-to-go function, thus reducing the computational cost of the dynamic programming procedure. As an example, the proposed procedure is applied to a water supply network scheduling problem.

Because there are many possibilities for the set of basic functions, parameters and operators used in the design of an adaptive-network-based fuzzy inference system (ANFTS), the search for the most suitable operators and functional blocks, together with their characterization and evaluation, is also an important topic in the field of neuro-fuzzy design. As shown in papers dealing with real applications, the designer has to select the operator to be used in each phase in the design of a neuro-fuzzy system, and this decision is usually taken in terms of the most common operations performed. Nevertheless, it is very important to determine which factors have the greatest influence on the behaviour and performance of the neuro-fuzzy system. Therefore, the designer should pay close attention to the phase in which the selection of the operator is most statistically significant. In this way, it is possible to obviate a detailed analysis of different configurations that lead to systems with very similar performance. In order to perform this analysis, an appropriate statistical tool has been used: the multifactorial analysis of the variance (ANOVA) which consists of a set of statistical techniques that enable the analysis and comparison of experiments, by describing the interactions and interrelations between either the quantitative or the qualitative variables of the neural network system. By applying this methodology to a great variety of neuro-fuzzy systems, it is possible to obtain general results about the most relevant factors defining the neural network design.

Describes a structure to create a RBF neural network. This structure has 4 main characteristics. The first one is that the special RBF network architecture uses regression weights to replace the constant weights normally used. These regression weights are assumed to be functions of input variables. The second characteristic is the normalization of the activation of the hidden neurons (weighted average) before aggregating the activations, which, as observed by various authors, produces better results than the classical weighted sum architecture. The third aspect is that a new type of nonlinear function is proposed, the pseudo-gaussian function (PGBF). With this, the neural system gains flexibility, as the neurons possess an activation field that does not necessarily have to be symmetric with respect to the centre or to the location of the neuron in the input space. In addition to this new structure, we propose, as the fourth and final feature, a sequential learning algorithm, which is able to adapt the structure of the network, with this, it is possible to create new hidden units and also to detect and remove inactive units.

In the synthesis of a fuzzy system two steps are generally employed: the identification of a structure and the optimization of the parameters defining it. The paper presents a methodology to automatically perform these two steps in conjunction using a three-phase approach to construct a fuzzy system from numerical data. Phase 1 outlines the membership functions and system rules for a specific structure, starting from a very simple initial topology. Phase 2 decides a new and more suitable topology with the information received from the previous step; it determines for which variable the number of fuzzy sets used to discretize the domain must be increased and where these new fuzzy sets should be located. This, in turn, decides in a dynamic way in which part of the input space the number of fuzzy rules should be increased. Phase 3 selects from the different structures obtained to construct a fuzzy system the one providing the best compromise between the accuracy of the approximation and the complexity of the rule set. The accuracy and complexity of the fuzzy system derived by the proposed self-organized fuzzy rule generation procedure (SOFRG) are studied for the problem of function approximation. Simulation results are compared with other methodologies such as artificial neural networks, neuro-fuzzy systems, and genetic algorithms.

In this paper, a systematic design is proposed to determine fuzzy system structure and learning its parameters, from a set of given training examples. In particular, two fundamental problems concerning fuzzy system modeling are addressed: 1) fuzzy rule parameter optimization and 2) the identification of system structure (i.e., the number of membership functions and fuzzy rules). A four-step approach to build a fuzzy system automatically is presented: Step 1 directly obtains the optimum fuzzy rules for a given membership function configuration. Step 2 optimizes the allocation of the membership functions and the conclusion of the rules, in order to achieve a better approximation. Step 3 determines a new and more suitable topology with the information derived from the approximation error distribution; it decides which variables should increase the number of membership functions. Finally, Step 4 determines which structure should be selected to approximate the function, from the possible configurations provided by the algorithm in the three previous steps. The results of applying this method to the problem of function approximation are presented and then compared with other methodologies proposed in the bibliography.

An evolutionary computation approach is used to learn online the rules that allow the processors in a parallel platform to cooperate by interchanging the local optima that they find while they concurrently explore different zones of the solution space. The cooperation of processors can greatly benefit the resolution of combinatorial optimization problems by decreasing their runtimes, by increasing the quality of the solutions obtained, or both. Moreover, as parallel computers are more and more accessible, the application of parallel processing to solve these problems becomes a practical and interesting alternative. As an example, a parallel optimization algorithm based on Boltzmann Machine has been used for a detailed description and evaluation of the proposed cooperation approach.

A general problem in model selection is to obtain the right parameters that make a model fit observed data. For a multilayer perceptron (MLP) trained with backpropagation (BP), this means finding the right hidden layer size, appropriate initial weights and learning parameters. The paper proposes a method (G-Prop-II) that attempts to solve that problem by combining a genetic algorithm (GA) and BP to train MLPs with a single hidden layer. The GA selects the initial weights and the learning rate of the network, and changes the number of neurons in the hidden layer through the application of specific genetic operators. G-Prop-II combines the advantages of the global search performed by the GA over the MLP parameter space and the local search of the BP algorithm. The application of the G-Prop-II algorithm to several real world and benchmark problems shows that MLPs evolved using G-Prop-II are smaller and achieve a higher level of generalization than other perceptron training algorithms, such as QuickPropagation or RPROP, and other evolutive algorithms, such as G-LVQ. It also shows some improvement over previous versions of the algorithm.

This paper shows a quantitative relation between the regularization techniques, the generalization ability, and the sensitivity of the Multilayer Perceptron (MLP) to input noise. Although many studies about these topics have been presented, in most cases only one of the problems is addressed, and only experimentally obtained evidence is provided to illustrate some kind of correlation between generalization, noise immunity and the use of regularization techniques to obtain a set of weights after training that provides the corresponding MLP with generalization ability and noise immunity. Here, a new measurement of noise immunity for a MLP is presented. This measurement, which is termed Mean Squared Sensitivity (MSS), explicitly evaluates the Mean Squared Error (MSE) degradation of a MLP when it is perturbed by input noise, and can be computed from the statistical sensitivities (previously proposed) of the output neurons. The MSS provides an accurate evaluation of the MLP performance loss when its inputs are perturbed by noise and can also be considered a measurement of the smoothness of the error surface with respect to the inputs. Thus, as the MSS can be used to evaluate the noise immunity or the generalization ability, it gives a criterion to select among different weight configurations that present a similar MSE after training.

Stimulus correlation and adaptive movement detection, among other tasks can be performed with VLSI general-purpose neurons that have controllable steady and transient responses. This paper presents experimental results of simple neural primitives based on the CMOS neuron approach described in [11]. Stimulus correlation experiments illustrate the well defined behavior of the CMOS approach. This basic primitive is used to implement motion detectors with adaptive capabilities that enable it to work efficiently in a wide velocity range.

The use of generalized Hopfield neural networks in designing the checking circuitry of a concurrent testable circuit is discussed. The aliasing probability, a measure for evaluating the performance of the checking circuitry, is provided. It is shown how, by using spectral techniques based on the Reed-Muller transform, the aliasing probability can be expressed as a function of the Reed-Muller coefficients. Therefore, obtaining the checking circuitry means selecting a set of Reed-Muller spectral coefficients, with fewer elements than a given bound, that minimizes the aliasing probability. To apply the neural networks to design the checking circuitry for concurrent testing, the aliasing probability has been used as an energy function, and the Hopfield neural network has been modified to have an associated energy function with any type of polynomial dependence on the processor states.>

Reed-Muller coefficients are used to generate a test-pattern selection procedure for detecting single stuck-at faults. This procedure is based on the heuristics deduced from the way in which the spectral coefficients are affected by such faults. The changes that the spectral coefficients undergo are also compared with the fault models used most frequently to model defects in combinational circuits. The proposed pattern-generating procedure does not need a simulation of the circuit for each of the possible stuck-at faults, and its complexity is proportional to the number of gates in the circuit. The proportionality constant increases exponentially as the number of inputs in the circuit increases. To evaluate the performance of the proposed method, its application to some benchmark circuits, including the ALU 74181, is presented.>