Search

This paper proposes an algorithm named Reverse Engineering Recursive Rule Extraction (RE-Re-RX) for symbolic rule extraction from neural network with mixed attributes. RE-Re-RX algorithm is an extension of the existing Recursive Rule Extraction (Re-RX) algorithm. Re-RX algorithm generates disjoint rules for continuous and discrete attributes. The algorithm first generates rules for discrete attributes. A rule for discrete attributes is further refined recursively if it does not produce satisfactory result. A rule is refined by generating rules with the discrete attributes (if present) that are not covered by the rule or else the process is terminated by generating rules with the continuous attributes (if present). The novelty of the proposed RE-Re-RX algorithm lies in generating rules for continuous attributes. Re-RX generates linear hyper plane for continuous attributes which may not be able to deal with the non-linearity present in data. To overcome this limitation RE-Re-RX algorithm generates simple rules for continuous attributes in the form of input data ranges and target. RE-Re-RX uses the concept of Rule Extraction by Reverse Engineering the NN (RxREN) algorithm in slightly different way to generate rules. RxREN only uses misclassified patterns, whereas RE-Re-RX uses both classified and misclassified patterns of each continuous attribute to calculate input data ranges for constructing rules. The proposed algorithm is validated with six benchmark datasets. The experimental results clearly show the superiority of the proposed algorithm to Re-RX. [ABSTRACT FROM AUTHOR]

The use of artificial neural networks for various problems has provided many benefits in various fields of research and engineering. Yet, depending on the problem, different architectures need to be developed and most of the time the design decision relies on a trial and error basis as well as on the experience of the developer. Many approaches have been investigated concerning the topology modelling, training algorithms, data processing. This paper proposes a novel automatic method for the search of a neural network architecture given a specific task. When selecting the best topology, our method allows the exploration of a multidimensional space of possible structures, including the choice of the number of neurons, the number of hidden layers, the types of synaptic connections, and the use of transfer functions. Whereas the backpropagation algorithm is being conventionally used in the field of neural networks, one of the known disadvantages of the technique represents the possibility of the method to reach saddle points or local minima, hence overfitting the output data. In this work, we introduce a novel strategy which is capable to generate a network topology with overfitting being avoided in the majority of the cases at affordable computational cost. In order to validate our method, we provide several numerical experiments and discuss the outcomes. [ABSTRACT FROM AUTHOR]

The use of computer-readable visual codes became common in our everyday life both in industrial environments and for private use. The reading process of visual codes consists of two steps, namely, localization and data decoding. In this paper we examine the localization step of visual codes using conventional and deep rectifier neural networks. They are also evaluated in the discrete cosine transform domain and shown to be efficient, which makes full decompression unnecessary for setups involving JPEG images. This approach is also efficient from a storage viewpoint and computation cost viewpoint, since camera hardware can provide a JPEG stream as output in many cases. The use of neural networks implemented on graphics processing unit allows real-time automatic code object localization. In our earlier studies, the proposed approach was evaluated on the most popular code type, quick response code, and some other 2D codes as well. Here, we also prove that deep rectifier networks are also suitable for 1D barcode localization and present extensive evaluation and comparison to state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

In this paper, the problem of exponential lag synchronization for a class of neural networks with mixed delays including discrete and distributed delays is investigated via adaptive intermittent control. Based on piecewise analytic method, some sufficient conditions for globally exponential lag synchronization are established through constructing a piecewise continuous auxiliary function. It is noted that both the control periods and the control widths in our adaptive intermittent control strategy are allowed to be nonidentical, which extends the scope of application of periodically intermittent control with fixed both control period and control width employed widely in previous works. Moreover, it is shown that the derived globally exponential lag synchronization criteria are related to the control rates rather than the control periods, which facilitates the choice of the control periods for practical problems. Finally, a numerical example is given to illustrate the correctness of the obtained theoretical results. [ABSTRACT FROM AUTHOR]

The scientific interest attracted by Spiking Neural Networks (SNN) has lead to the development of tools for the simulation and study of neuronal dynamics ranging from phenomenological models to the more sophisticated and biologically accurate Hodgkin-and-Huxley-based and multi-compartmental models. However, despite the multiple features offered by neural modelling tools, their integration with environments for the simulation of robots and agents can be challenging and time consuming. The implementation of artificial neural circuits to control robots generally involves the following tasks: (1) understanding the simulation tools, (2) creating the neural circuit in the neural simulator, (3) linking the simulated neural circuit with the environment of the agent and (4) programming the appropriate interface in the robot or agent to use the neural controller. The accomplishment of the above-mentioned tasks can be challenging, especially for undergraduate students or novice researchers. This paper presents an alternative tool which facilitates the simulation of simple SNN circuits using the multi-agent simulation and the programming environment Netlogo (educational software that simplifies the study and experimentation of complex systems). The engine proposed and implemented in Netlogo for the simulation of a functional model of SNN is a simplification of integrate and fire (I&F) models. The characteristics of the engine (including neuronal dynamics, STDP learning and synaptic delay) are demonstrated through the implementation of an agent representing an artificial insect controlled by a simple neural circuit. The setup of the experiment and its outcomes are described in this work. [ABSTRACT FROM AUTHOR]

To find a route with good stability and less cost is a hot issue because of MANET's mobility. AODV is one of the most widely used routing protocols in MANET because of its wide application, good performance and expansion. However, AODV is only an optional route instead of an optimized one. In this paper, continuous Hopfield Neural Networks is used to optimize the route to seek an optimal or nearly-optimal route, which can improve the usability and survivability of MANET. The simulation results show that CHNN-AODV is more effective and advantageous than AODV in the measurement of packet receiving rate, end-to-end average delay and routing recovery frequency. [ABSTRACT FROM AUTHOR]

The paper presents the application of the artificial neural network algorithm for the retrieval of total precipitable water in the atmosphere over water and land from the measurements of MTVZA-GYa microwave radiometer on board the Meteor-M No. 2-2 satellite. Satellite-based estimates of total precipitable water were compared with radiosonde and AERONET data, as well as with the ECMWF numerical weather prediction model output. According to the comparison, the root-mean-square error (RMSE) does not exceed 4.5 mm for radiosonde data and is less than 4 mm for the ECMWF and AERONET data. The best accuracy is provided over water with the RMSE not exceeding 3 mm. The total precipitable water estimates retrieved from MTVZA-GYa and NOAA-20/ATMS radiometer data are consistent over water, while the MTVZA-GYa based estimates are more accurate over land. [ABSTRACT FROM AUTHOR]

The tuning of the viscosity of polyurethane (PU) during its polymerization enables its use in different applications including the liquid composite molding (LCM). In this work, the rheological behavior during polymerization of PU from castor oil (CO), polyether (PE), and a mixture of both polyols (COPE, 50/50 wt%) was investigated using oscillatory rheometry at different isotherms (40, 50, 60, 70, and 80 °C). Approaches based on artificial neural networks (ANNs) and response surface methodology (RSM) were used for data analysis and to obtain prediction tools related to the PU polymerization. This approach can contribute to experimentally estimate new formulations to obtain information that kinetic models based on conversion rates are unable to provide. The polymerized PU(CO), PU(COPE), and PU(PE) compositions showed flexible phase Tg of 15, − 13, and − 30 °C, respectively. Due to the greater reactivity of the PU(CO), its viscosity increased faster than that of PU(PE). The balance of properties obtained with the mixtures of polyols resulted in viscosity values of 300, 500, and 1000 mPa s, which are suitable for the LCM. This approach shows the combination of ANN and RSM can be advantageous to complement the kinetics analysis of PU resins with tunable properties for the LCM. Highlights: Rheokinetics of PUs analyzed using oscillatory rheometry. Experimental data examined with neural networks and response surface methodology. Successful utilization of ANN and RSM for predicting PU polymerization. Combined ANN and RSM enhance kinetics analysis of PU resins. This approach reveals viscosity values suitable for composite liquid molding. [ABSTRACT FROM AUTHOR]

Deep neural networks have shown remarkable performance on a wide range of classification tasks and applications. However, the large model size and the enormous size of the training dataset make the training process slow and often limited by the computing resources. To overcome this limitation, distributed training can be used to accelerate the process by utilizing multiple devices for a single model training. In this work, we evaluate the performance of Microsoft DeepSpeed, a distributed training library, on image classification tasks by comparing the performance of 108 trained neural networks in 27 unique settings. Our experimental results suggest that DeepSpeed may provide limited benefits for simpler learning tasks (e.g. smaller neural network models or simpler datasets). On the other hand, for more complex learning tasks, DeepSpeed can provide up to 8× faster training with possible performance improvement. Our study contributes to a better understanding of the capabilities and limitations of the DeepSpeed library, providing insights into when and where it may be most beneficial to use in image classification settings. [ABSTRACT FROM AUTHOR]

The present paper covers the issue of training of an artificial neural network in an intelligent diagnostic system whose purpose is to evaluate repairable technical objects. The structure of the diagnostic system was characterized, and the measurement and diagnostic subsystems were described. An artificial neural network is an important element in an intelligent diagnostic subsystem. The structure, the algorithm, the organization of a neural network and the basic relations that describe its work were presented. The information presented in the form of the vectors of diagnostic signals, and their standard vectors constitute the primary information base used in 'DIAG' computer program. Training of an artificial neural network is an important aspect that is presented in the paper. The issue concerning these problems is not presented in the literature. Training of a network was presented on the grounds of teaching vectors, which are determined in a diagnostic system in the process of a simulation of a specific state in the object examined. An example of training of a network was presented in a diagnostic system which evaluates a control system of the operation of a car engine. Appropriate connections were presented for the purpose of a qualitative assessment of the training process of a neural network. [ABSTRACT FROM AUTHOR]

This paper deals with the H control problem of neural networks with time-varying delays. The system under consideration is subject to time-varying delays and various activation functions. Based on constructing some suitable Lyapunov-Krasovskii functionals, we establish new sufficient conditions for H control for two cases of time-varying delays: (1) the delays are differentiable and have an upper bound of the delay-derivatives and (2) the delays are bounded but not necessary to be differentiable. The derived conditions are formulated in terms of linear matrix inequalities, which allow simultaneous computation of two bounds that characterize the exponential stability rate of the solution. Numerical examples are given to illustrate the effectiveness of our results. [ABSTRACT FROM AUTHOR]

Semi-supervised clustering is gaining importance these days since neither supervised nor unsupervised learning methods in a stand-alone manner provide satisfactory results. Existing semi-supervised clustering techniques are mostly based on pair-wise constraints, which could be misleading. These semi-supervised clustering algorithms also fail to address the problem of dealing with attributes having different weights. In most of the real-life applications, all attributes do not have equal importance and hence same weights cannot be assigned for each attribute. In this paper, a novel distance-based semi-supervised clustering algorithm has been proposed, which uses functional link neural network (FLNN) for finding weights for attributes with small amount of labeled data for further use in parametric Minkowski's model for clustering. In FLNN, the nonlinearity is captured by enhancing the input using orthonormal basis functions. The effectiveness of the approach has been illustrated over a number of datasets taken from UCI machine learning repository. Comparative performance evaluation demonstrates that the proposed approach outperforms the existing semi-supervised clustering algorithms. The proposed approach has also been successfully used to cluster the crime locations and to find crime hot spots in India on the data provided by National Crime Records Bureau (NCRB). [ABSTRACT FROM AUTHOR]

The mixed use of different shapes of radial basis functions (RBFs) in radial basis functions neural networks (RBFNNs) is investigated in this paper. For this purpose, we propose the use of a generalised version of the standard RBFNN, based on the generalised Gaussian distribution. The generalised radial basis function (GRBF) proposed in this paper is able to reproduce other different radial basis functions (RBFs) by changing a real parameter τ. In the proposed methodology, a hybrid evolutionary algorithm (HEA) is employed to estimate the number of hidden neuron, the centres, type and width of each RBF associated with each radial unit. In order to test the performance of the proposed methodology, an experimental study is presented with 20 datasets from the UCI repository. The GRBF neural network (GRBFNN) was compared to RBFNNs with Gaussian, Cauchy and inverse multiquadratic RBFs in the hidden layer and to other classifiers, including different RBFNN design methods, support vector machines (SVMs), a sparse probabilistic classifier (sparse multinominal logistic regression, SMLR) and other non-sparse (but regularised) probabilistic classifiers (regularised multinominal logistic regression, RMLR). The GRBFNN models were found to be better than the alternative RBFNNs for almost all datasets, producing the highest mean accuracy rank. [ABSTRACT FROM AUTHOR]

This paper deals with the robust stability problem of uncertain stochastic neural networks of neutral-type with interval time-varying delays. The uncertainties under consideration are norm-bounded, and the delay is assumed to be time-varying and belongs to a given interval. By using the Lyapunov-Krasovskill functional method and the linear matrix inequality (LMI) technique, the novel stability criteria are derived in terms of LMI. Finally, numerical examples are provided to demonstrate the effectiveness of the proposed criteria. [ABSTRACT FROM AUTHOR]

In this paper, we present a multiuser detection technique based on artificial neural network (NN) for synchronous multicarrier code division multiple access systems over Rayleigh fading channels. To test the robustness of the proposed receiver, also the effect of power control problem is studied with a comparative manner. Bit error rate (BER) performance of the NN based receiver is compared with the single user bound and conventional receivers. Although the BER performance of the conventional receiver degrades as the number of the users and power level differences among the users increase, as a decision structure, neural network based receiver gives closer BER performance to the single user bound. [ABSTRACT FROM AUTHOR]

The analysis of blood cells in microscope images can provide useful information concerning the health of patients; however, manual classification of blood cells is time-consuming and susceptible to error due to the different morphological features of the cells. Therefore, a fast and automated method for identifying the different blood cells is required. In this paper, we investigate the use of different neural network models for the purpose of cell identification. The neural models are based on the back propagation learning algorithm and differ in design according to the way data features are extracted from the cell microscopic images. Three different topologies of neural networks are investigated, and a comparison between these models is drawn. Experimental results suggest that the proposed method performs well in identifying blood cell types regardless of their irregular shapes, sizes, and orientation. [ABSTRACT FROM AUTHOR]

Fuzzy systems have gained more and more attention from researchers and practitioners of various fields. In such systems, the output represented by a fuzzy set sometimes needs to be transformed into a scalar value, and this task is known as the defuzzification process. Several analytic methods have been proposed for this problem, but lately, the neural network approach has been used for this purpose. When employed as defuzzifiers, a neural network is called a defuzzification neural network. In this paper, some preliminary results on properties of such defuzzification networks will be reported. [ABSTRACT FROM AUTHOR]

In this paper we discuss the existence and global attractivity of k-pseudo almost automorphic sequence solution of a model of bidirectional cellular neural networks. We consider the corresponding difference equation analogue of the model system using suitable discretization method and obtain certain conditions for the existence of solution. The k-pseudo almost automorphic sequence solutions generalize the results of pseudo almost periodic, almost periodic and almost automorphic sequences solutions. Moreover the results proved in this paper are new and compliment the existing one. [ABSTRACT FROM AUTHOR]

There has been a constant endeavor towards improving the available circuit design automation tools to match technological advancements in the electronic industry. However, inadequate research efforts in the analog domain are holding back the exploitation of advanced technologies. A dearth of design expertise in the analog domain is the principal driving force for the growth of Design Automation (DA) tools. Transistor sizing is one of the most crucial steps in the analog IC design. In this paper, we put forward a new computer aided design framework for the sizing of transistors in MOS Integrated Circuit (IC) amplifiers by incorporating powerful modeling capabilities of Artificial Neural Networks (ANN). ANNs have proven to be efficient and accurate modeling tools in several applications. The proposed tool is capable of directly computing transistor related design parameters, of the MOS IC amplifier and associated peripheral circuitry. The proposed tool thus avoids several time-consuming simulations and/or tuning runs at the very bottom level of analog IC amplifier implementation, using a given CMOS process. It also reduces manual intervention in the design process, thus enhancing the automation of the design process. This paper presents design examples of several analog IC functional modules that are developed and verified successfully. [ABSTRACT FROM AUTHOR]

This paper proposes a self-constructing fuzzy neural network-based decision feedback equalizer (SCFNN DFE). An online learning algorithm containing the structure and parameter learning phases is employed in training the SCFNN DFE. Specifically, the feedforward input vector classification and a gradient-descent method are both used in this online learning algorithm. We show by simulations that the proposed SCFNN DFE offers improvement compared to the traditional DFE methods in the presence of frequency offset and phase noise. [ABSTRACT FROM AUTHOR]

This paper presents two new approaches for constructing an ensemble of neural networks (NN) using coevolution and the artificial immune system (AIS). These approaches are extensions of the CLONal Selection Algorithm for building ENSembles (CLONENS) algorithm. An explicit diversity promotion technique was added to CLONENS and a novel coevolutionary approach to build neural ensembles is introduced, whereby two populations representing the gates and the individual NN are coevolved. The former population is responsible for defining the ensemble size and selecting the members of the ensemble. This population is evolved using the differential evolution algorithm. The latter population supplies the best individuals for building the ensemble, which is evolved by AIS. Results show that it is possible to automatically define the ensemble size being also possible to find smaller ensembles with good generalization performance on the tested benchmark regression problems. More interestingly, the use of the diversity measure during the evolutionary process did not necessarily improve generalization. In this case, diverse ensembles may be found using only implicit diversity promotion techniques. [ABSTRACT FROM AUTHOR]

This paper proposes a novel steam valving controller using radial basis function (RBF) networks-based approximate model method. Approximate model method is a kind of direct linearization approach that is derived based on the approximation of the plant's input-output model via Taylor expansion. RBF networks are used to identify the plant to implement the approximate model control law. In order to improve the performance of the approximate model controller, RBF networks weights are adjusted online using BP algorithms with an adaptive learning rate. Several simulations results demonstrate the effectiveness of the proposed controller for team valving control. [ABSTRACT FROM AUTHOR]

critical issue of Neural Network based large-scale data mining algorithms is how to speed up their learning algorithm. This problem is particularly challenging for Error Back-Propagation (EBP) algorithm in Multi-Layered Perceptron (MLP) Neural Networks due to their significant applications in many scientific and engineering problems. In this paper, we propose an Adaptive Variable Learning Rate EBP algorithm to attack the challenging problem of reducing the convergence time in an EBP algorithm, aiming to have a high-speed convergence in comparison with standard EBP algorithm. The idea is inspired from adaptive filtering, which leaded us into two semi-similar methods of calculating the learning rate. Mathematical analysis of AVLR-EBP algorithm confirms its convergence property. The AVLR-EBP algorithm is utilized for data classification applications. Simulation results on many well-known data sets shall demonstrate that this algorithm reaches to a considerable reduction in convergence time in comparison to the standard EBP algorithm. The proposed algorithm, in classifying the IRIS, Wine, Breast Cancer, Semeion and SPECT Heart datasets shows a reduction of the learning epochs relative to the standard EBP algorithm. [ABSTRACT FROM AUTHOR]

In this paper, we present a review of various computational experiments concerning neural network (NN) models developed for regional employment forecasting. NNs are nowadays widely used in several fields because of their flexible specification structure. A series of NN experiments is presented in the paper, using two data sets on German NUTS-3 districts. Individual forecasts are computed by our models for each district in order to answer the following question: How relevant are NN parameters in comparison to NN structure? Comprehensive testing of these parameters is limited in the literature. Building on different specifications of NN models-in terms of explanatory variables and NN structures-we propose a systematic choice of NN learning parameters and internal functions by means of a sensitivity analysis. Our results show that different combinations of NN parameters provide significantly varying statistical performance and forecasting power. Finally, we note that the sets of parameters chosen for a given model specification cannot be light-heartedly applied to different or more complex models. [ABSTRACT FROM AUTHOR]

Contemporary image recognition schemes either rely on single-feature recognition or focus on solving multi-feature recognition using complex computational approaches. Furthermore these approaches tend to be of tightly-coupled nature, thus not readily deployable within computational networks. Distributed Hierarchical Graph Neuron (DHGN) is a distributed single-cycle learning pattern recognition algorithm that can scale from coarse-grained to fine-grained networks and it has comparable accuracy to contemporary image recognition schemes. In this paper, we present an implementation of DHGN that works for multi-feature recognition of images. Our scheme is able to disseminate recognition of each feature within an image to a separate computational subnetwork. Thereby allowing a number of features being analysed simultaneously using a uniform recognition process. We have conducted tests on a collection of greyscale facial images. The results show that our approach produces high recognition accuracy through a simple distributed process. Furthermore, our approach implements single-cycle learning known as collaborative-comparison learning where new patterns are continuously stored using collaborative approach without affecting previously stored patterns. Our proposed scheme demonstrates higher classification accuracy in comparison with Back-Propagation Neural Network for multi-class images. [ABSTRACT FROM AUTHOR]

Road safety performance indicators are comprehensible tools that provide a better understanding of current safety conditions and can be used to monitor the effect of policy interventions. New insights can be gained in case one road safety index is composed of all risk indicators. The overall safety performance can then be evaluated, and countries ranked. In this paper, a promising structure of neural networks based on decision rules generated by rough sets—is proposed to develop an overall road safety index. This novel hybrid system integrates the ability of neural networks on self-learning and that of rough sets on automatically transforming data into knowledge. By means of simulation, optimal weights are assigned to seven road safety performance indicators. The ranking of 21 European countries in terms of their road safety index scores is compared to a ranking based on the number of road fatalities per million inhabitants. Evaluation results imply the feasibility of this intelligent decision support system and valuable predictive power for the road safety indicators context. [ABSTRACT FROM AUTHOR]

In this paper we continue our investigation of the analogical neural network, by introducing and studying its replica symmetric approximation in the absence of external fields. Bridging the neural network to a bipartite spin-glass, we introduce and apply a new interpolation scheme to its free energy, that naturally extends the interpolation via cavity fields or stochastic perturbations from the usual spin glass case to these models. While our methods allow the formulation of a fully broken replica symmetry scheme, in this paper we limit ourselves to the replica symmetric case, in order to give the basic essence of our interpolation method. The order parameters in this case are given by the assumed averages of the overlaps for the original spin variables, and for the new Gaussian variables. As a result, we obtain the free energy of the system as a sum rule, which, at least at the replica symmetric level, can be solved exactly, through a self-consistent mini-max variational principle. The so gained replica symmetric approximation turns out to be exactly correct in the ergodic region, where it coincides with the annealed expression for the free energy, and in the low density limit of stored patterns. Moreover, in the spin glass limit it gives the correct expression for the replica symmetric approximation in this case. We calculate also the entropy density in the low temperature region, where we find that it becomes negative, as expected for this kind of approximation. Interestingly, in contrast with the case where the stored patterns are digital, no phase transition is found in the low temperature limit, as a function of the density of stored patterns. [ABSTRACT FROM AUTHOR]

'Particularism' and 'generalism' refer to families of positions in the philosophy of moral reasoning, with the former playing down the importance of principles, rules or standards, and the latter stressing their importance. Part of the debate has taken an empirical turn, and this turn has implications for AI research and the philosophy of cognitive modeling. In this paper, Jonathan Dancy's approach to particularism (arguably one of the best known and most radical approaches) is questioned both on logical and empirical grounds. Doubts are raised over whether Dancy's brand of particularism can adequately explain the graded nature of similarity assessments in analogical arguments. Also, simple recurrent neural network models of moral case classification are presented and discussed. This is done to raise concerns about Dancy's suggestion that neural networks can help us to understand how we could classify situations in a way that is compatible with his particularism. Throughout, the idea of a surveyable standard-one with restricted length and complexity-plays a key role. Analogical arguments are taken to involve multidimensional similarity assessments, and surveyable contributory standards are taken to be attempts to articulate the dimensions of similarity that may exist between cases. This work will be of relevance both to those who have interests in computationally modeling human moral cognition and to those who are interested in how such models may or may not improve our philosophical understanding of such cognition. [ABSTRACT FROM AUTHOR]

In this paper, the dynamic behaviors of a class of neural networks with time-varying delays are investigated. Some less weak sufficient conditions based on p-norm and ∞-norm are obtained to guarantee the existence, uniqueness of the equilibrium point for the addressed neural networks without impulsive control by applying homeomorphism theory. And then, by utilizing inequality technique, Lyapunov functional method and the analysis method, some new and useful criteria of the globally exponential stability with respect to the equilibrium point under impulsive control we assumed are derived based on p-norm and ∞-norm, respectively. Finally, an example with simulation is given to show the effectiveness of the obtained results. [ABSTRACT FROM AUTHOR]

Advanced monitoring systems enable integration of data-driven algorithms for various tasks, for e.g., control, decision support, fault detection and isolation (FDI), etc. Due to improvement of monitoring systems, statistical or other computational methods can be implemented to real industrial systems. Algorithms which rely on process history data sets are promising for real-time operation especially for online process monitoring tasks, e.g., FDI. However, a reliable FDI system should be robust to uncertainties and small process deviations, thus, false alarms can be avoided. To achieve this, a good model for comparison between process and model is needed and for easier FDI implementation, the model has to be derived directly from process history data. In such cases, model-based FDI approaches are not very practical. In this paper a nonlinear statistical multivariate method (nonlinear principal component analysis) was used for modeling, and realized with auto-associative artificial neural network (AANN). A Taguchi design of experiments (DoE) technique was used and compared with a classic approach, where according to the analysis best AANN model structure was chosen for nonlinear model. Parameters that are important for neural network’s performance have been included into a joint orthogonal array to consider interactions between noise and control process variables. Results are compared to AANN design recommendations by other authors, where obtained nonlinear model was designed for reliable fault detection of very small faults under closed-loop conditions. By using Taguchi DoE robust design on AANN, an improved and reliable FDI scheme was achieved even in case of small faults introduced to the system. The accuracy and performance of AANN and FDI scheme were tested by experiments carried out on a real laboratory hydraulic system, to validate the proposed design for industrial cases. [ABSTRACT FROM AUTHOR]

The paper studies the design and analysis of a neural adaptive control strategy for a class of square nonlinear bioprocesses with incompletely known and time-varying dynamics. In fact, an adaptive controller based on a dynamical neural network used as a model of the unknown plant is developed. The neural controller design is achieved by using an input–output feedback linearization technique. The adaptation laws of neural network weights are derived from a Lyapunov stability property of the closed-loop system. The convergence of the system tracking error to zero is guaranteed without the need of network weights convergence. The resulted control method is applied in a depollution control problem in the case of a wastewater treatment bioprocess, belonging to the square nonlinear class, for which kinetic dynamics are strongly nonlinear, time varying and not exactly known. [ABSTRACT FROM AUTHOR]

In modern world of today where air traffic is continuously increasing and available space at the airports remains finite, there is a problem of safe docking of aircraft. The problem needs to be solved to ensure safe and smooth movement of aircraft, passengers and crew while making optimum utilization of available ground space. Without such systems having in place, accidents keep occurring due to human judgment errors. These accidents are causing loss of material costs and human injury. The importance of Video-based Docking Systems is continuously increasing due to the challenges of current and upcoming traffic demands of future. This paper evaluates two neural networks architectures for recognition of civil airliners in a Video docking system. The networks compared are feedforward neural network and probabilistic neural network. The results are compared by presenting data to neural networks while deforming the shape, adding noise and partly occluding the shape and presenting those angles for which network was not trained. [ABSTRACT FROM AUTHOR]

This paper presents a method to control an operation process of a complex technical object, for example a car engine, with the use of trivalent diagnostic information. Also, a general diagram of the complex technical object was presented, and its internal structure was described. A diagnostic analysis was conducted, as a result of which sets of the functional elements of the object and its diagnostic signals were determined. Also, the methodology of the diagnostic examination of the technical system was presented. The result was a functional and diagnostic model, which constituted the basis for initial diagnostic information, which is provided by the sets of information concerning the elements of the basic modules and their output signals. The article also covers a diagnostic system which uses a DIAG computer programme for the recognition of the states of technical objects in trivalent logics. A programme was presented and described in an analytical form for diagnosis through an artificial neural network (ANN), which recognises the states of reparable technical objects in trivalent logics. The final results obtained from the computations conducted by the DIAG programme are presented in the table of the states of the object. [ABSTRACT FROM AUTHOR]

The paper formulates the problem and considers the vector optimization of the architecture of a neural network classifier. The scalar convolution of criteria according to a nonlinear tradeoff scheme is proposed as a criterion function. Search methods of optimization with discrete arguments are used. An example of a neural network text classifier is given. [ABSTRACT FROM AUTHOR]

This paper investigates the effectiveness of using the Contract Net Protocol, an auction type system, for controlling task allocation among a group of robots, and presents and evaluates a strategy of using Artificial Neural Networks to formulate adaptive bids within the framework of the Contract Net Protocol. The robots were used in a foraging environment and showed that excellent communication among robots leads to a need for a social control mechanism for managing the robots, such as the Contract Net Protocol. The experiments also confirmed that a moderate benefit can be gained by using adaptive bidding within the framework of the Contract Net Protocol. [ABSTRACT FROM AUTHOR]

A novel hierarchical hybrid neural model to the problem of long-term load forecasting is proposed in this paper. The neural model is made up of two self-organizing map nets—one on top of the other—and a single-layer perceptron. It has application into domains which require time series analysis. The model is compared to a mutilated architecture of it, and to a multilayer perceptron. The hierarchical, the mutilated, and the multilayer perceptron models are trained and assessed on load data extracted from a North-American electric utility. They are required to predict either once every week or once every month the electric peak-load during the next two years. The results from the experiments show that the performance of HHNM on long-term load forecasts is better than that of the mutilated model, and much better than that of the MLP model. [ABSTRACT FROM AUTHOR]

Compared to normal learning algorithms, for example backpropagation, the optimal bounded ellipsoid (OBE) algorithm has some better properties, such as faster convergence, since it has a similar structure as Kalman filter. OBE has some advantages over Kalman filter training, the noise is not required to be Guassian. In this paper OBE algorithm is applied in training the weights of the feedforward neural network for nonlinear system identification. Both hidden layers and output layers can be updated. From a dynamic system point of view, such training can be useful for all neural network applications requiring real-time updating of the weights. Two simulations give the effectiveness of the suggested algorithm. [ABSTRACT FROM AUTHOR]

In our attempts to model human intelligence by mimicking the brain structure and function, we overlook an important aspect in human cognition, which is the emotional factor. It may currently sound unthinkable to have emotional machines; however, it is possible to simulate certain artificial emotions with the aim of improving machine learning. This paper investigates the efficiency of an emotional neural network, which uses a modified back propagation learning algorithm. The modifed algorithm, namely the emotional BP learning algorithm, has two emotional parameters, anxiety and confidence, that are modeled during machine learning and decision making. The emotional neural network will be implemented to a facial recognition problem using images of faces with different orientations and contrast levels, and its performance will be compared to that of a conventional neural network. Experimental results suggest that artificial emotions can be successfully modeled and efficiciently implemented to improve neural networks learning and generaliztion. [ABSTRACT FROM AUTHOR]

Color is a powerful descriptor that often simplifies object extraction and identification, and many computer vision systems use color to aid object recognition. However, image colors strongly depend on lighting geometry (direction and intensity of light source) and illuminant color (spectral power distribution). Either small variation in the intensity or the change of scene illumination can dramatically make object color changed. To overcome the lighting dependency problem, a color constancy or normalization algorithm should be used for preprocessing. This paper presents a novel approach to performing color normalization. A nonlinear mapping function is estimated using a neural network. Once the mapping function is found accurately, an image under unknown illumination may be transformed to the image under the predetermined illumination, which will be useful for color image processing. Three groups of experiments were conducted. In our experiments, images are processed by various neural networks and the performance is boosted by using a committee machine, and then the mapping errors are estimated and the results are compared with those of other algorithms. The experimental results demonstrate that the performance of the proposed method is superior to that of other color normalization algorithms. [ABSTRACT FROM AUTHOR]

In this paper, we consider the problem of exponential stability for recurrent neural networks with multiple time-varying delays and reaction-diffusion terms. The activation functions are supposed to be bounded and globally Lipschitz continuous. By means of Lyapunov functionals, sufficient conditions are derived, which guarantee global exponential stability of the delayed neural network. Finally, a numerical example is given to show the correctness of our analysis. [ABSTRACT FROM AUTHOR]

In this paper, the entropy concept has been utilized to characterize the uncertainty of the tracking error for nonlinear ARMA stochastic systems over a communication network, where time delays in the communication channels are of random nature. A recursive optimization solution has been developed. In addition, an alternative algorithm is also proposed based on the probability density function of the tracking error, which is estimated by a neural network. Finally, a simulation example is given to illustrate the efficiency and feasibility of the proposed approach. [ABSTRACT FROM AUTHOR]

Optimizing the traffic signal control has an essential impact on intersections efficiency in urban transportation. This paper presents a two-stage method for intersection signal timing control. First, the traffic volume is predicted using a neuro-fuzzy network called Adaptive neuro-fuzzy inference system (ANFIS). The inputs of this network include two-dimensional, hourly and daily, traffic volume correlations. In the second stage, appropriate signal cycle and optimized timing of each phase of the signal are estimated using a combination of Self Organizing and Hopfield neural networks. The energy function of the Hopfield network is based on a traffic model derived by queuing analysis. The performance of the proposed method has been evaluated for real data. The two-dimensional correlation presents superior performance compared to hourly traffic correlation. The evaluation of proposed overall method shows considerable intersection throughput improvement comparing to the results taken form Synchro software. [ABSTRACT FROM AUTHOR]

Optical Character Recognition (OCR) systems have been effectively developed for the recognition of printed characters of non-Indian languages. Efforts are on the way for the development of efficient OCR systems for Indian languages, especially for Kannada, a popular South Indian language. We present in this paper an OCR system developed for the recognition of basic characters (vowels and consonants) in printed Kannada text, which can handle different font sizes and font types. Hu's invariant moments and Zernike moments that have been progressively used in pattern recognition are used in our system to extract the features of printed Kannada characters. Neural classifiers have been effectively used for the classification of characters based on moment features. An encouraging recognition rate of 96⋅8% has been obtained. The system methodology can be extended for the recognition of other south Indian languages, especially for Telugu. [ABSTRACT FROM AUTHOR]

Multi-variable generalized predictive control algorithm has obtained great success in process industries. However, it suffers from a high computational cost because the multi-stage optimization approach in the algorithm is time-consuming when constraints of the control system are considered. In this paper, a dual neural network is employed to deal with the multi-stage optimization problem, and bounded constraints on the input and output signals of the control system are taken into account. The dual neural network has many favorable features such as simple structure, rapid execution, and easy implementation. Therefore, the computation efficiency, in comparison with the consecutive executions of numerical algorithms on digital computers, is increased dramatically. In addition, the dual network model can yield the exact optimum values of future control signals while many other neural networks only obtain the approximate optimal solutions. Hence the multi-variable generalized predictive control algorithm based on the dual neural network is suitable for industrial applications with the real-time computation requirement. Simulation examples are given to demonstrate the efficiency of the proposed approach. [ABSTRACT FROM AUTHOR]

This paper describes the use of artificial intelligence-based techniques for detecting and isolating sensor failures in a turbojet engine. Specifically, three artificial intelligence (Al) techniques are employed: artificial neural networks (NNs), statistical expectations, and Bayesian belief networks (BBNs). These techniques are combined into an overall system that is capable of distinguishing between sensor failure and engine failure¿a critical capability in the operation of turbojet engines. The turbojet engine used in this study is an SR-30 developed by Turbine Technologies. Initially, NNs were designed and trained to recognize sensor failure in the engine. The increased random noise output from failing sensors was used as the key indicator. Next, a Bayesian statistical method was used to recognize sensor failure based on the bias error occurring in the sensors. Finally, a BBN was developed to interpret the results of the jNN and statistical evaluations. The BBN determines whether single or multiple sensor failures signify engine failure, or whether sensor failures represent separate, unrelated incidences. The BBN algorithm is also used to distinguish between bias and noise errors on sensors used to monitor turbojet performance. The overall system is demonstrated to work equally well during start-up and main-stage operation of the engine. Results show that the method can efficiently detect and isolate single or multiple sensor failures within this dynamic environment. [ABSTRACT FROM AUTHOR]

In this paper, we introduce a concept of advanced self-organizing polynomial neural network (Adv_SOPNN). The SOPNN is a flexible neural architecture whose structure is developed through a modeling process. But the SOPNN has a fatal drawback; it cannot be constructed for nonlinear systems with few input variables. To relax this limitation of the conventional SOPNN, we combine a fuzzy system and neural networks with the SOPNN. Input variables are partitioned into several subspaces by the fuzzy system or neural network, and these subspaces are utilized as new input variables to the SOPNN architecture. Two types of the advanced SOPNN are obtained by combining not only the fuzzy rules of a fuzzy system with SOPNN but also the nodes in a hidden layer of neural networks with SOPNN into one methodology. The proposed method is applied to the nonlinear system with two inputs, which cannot be identified by conventional SOPNN to show the performance of the advanced SOPNN. The results show that the proposed method is efficient for systems with limited data set and a few input variables and much more accurate than other modeling methods with respect to identification error. [ABSTRACT FROM AUTHOR]

In this paper, an image restoration algorithm is proposed to identify nonlinear and noncausal blur funclon using artificial neural networks. Image and degradation processes include both linear and nonlinear phenomena. The proposed neural network model, which combines an adaptive auto-associative network with a random Gaussian process, is used to restore the blurred image and blur function, simultaneously. The noisy and blurred images are modeled as nonlinear continuous associative networks. The auto-associative part determines the image model coefficients and the hetero-associative part determines the blur function of the image degradation process. The self-organization like structure of the proposed neural network provides the potential solution of the blind image restoration problem. The estimation and restoration are implemented by using an iterative gradient based algorithm to minimize the error function. [ABSTRACT FROM AUTHOR]

Autonomous neural network systems typically require fast learning and good generalization performance, and there is potentially a trade-off between the two. The use of evolutionary techniques to improve the learning abilities of neural network systems is now widespread. However, there are a range of different evolutionary approaches that could be applied, and no systematic investigation has been carried out to find which work best. In this paper, such an investigation is presented, and it is shown that a range of evolutionary techniques can generate high performance networks, but they often lead to unwanted side effects, such as occasional instances of very poor performance. The nature of these problems are explored further, and it is shown how the evolution of age dependent plasticities and/or the use of ensemble techniques can alleviate them. A range of techniques are thus identified, with differing properties, that can be matched to the specific requirements of each application. [ABSTRACT FROM AUTHOR]

This paper presents an intelligent process planning system using STEP features (ST-FeatCAPP) for prismatic parts. The system maps a STEP AP224 XML data file, without using a complex feature recognition process, and produces the corresponding machining operations to generate the process plan and corresponding STEP-NC in XML format. It carries out several stages of process planning such as operations selection, tool selection, machining parameters determination, machine tools selection and setup planning. A hybrid approach of most recent techniques (neural networks, fuzzy logic and rule-based) of artificial intelligence is used as the inference engine of the developed system. An object-oriented approach is used in the definition and implementation of the system. An example part is tested and the corresponding process plan is presented to demonstrate and verify the proposed CAPP system. The paper thus suggests a new feature-based intelligent CAPP system for avoiding complex feature recognition and knowledge acquisition problems. [ABSTRACT FROM AUTHOR]

This paper investigates an intelligent system that selects dispatching rules to apply locally for each machine in a job shop. Randomly generated problems are scheduled using optimal permutations of three different dispatching rules on five machines. A neural network is then trained to associate between a statistical characterization of the job mix in each of these problems, with the best combination of dispatching rules to use. Once trained, the neural network is able to recommend for new problems a dispatching rule to use on each machine. Two networks are trained separately for minimizing makespan and the mean flowtime in the job shop. Test results show that the combinations of dispatching rules suggested by the trained networks produce better results, for both objectives, than the alternative of using a single rule common to all machines. [ABSTRACT FROM AUTHOR]