Search

The paper illustrates the process of generating synthetic data using diffusion neural networks for solving popular computer vision tasks in urban applications: traffic vehicles detection and classification. The result of this study demonstrated new approach for synthetic datasets creation using modern generative neural networks. To elaborate this approach, experiments were performed to generate synthetic dataset to solve the above basic computer vision problem in urban applications using the most advanced diffusion neural network models (Kandinsky 2.2). Experiments were performed on these datasets to train deep neural networks to solve the object detection problem (YOLOv5). The results of testing the trained detectors on a specially selected validation dataset showed the potential viability of a synthetic dataset generation approach using diffusion neural network models. However, full-fledged use of this approach to generate synthetic datasets that can be used in training deep neural networks of computer vision in production is accompanied by some difficulties, namely high difference between domains of real data and generated data, high labor costs of tuning trained diffusion models to achieve the highest quality of generation, etc. [ABSTRACT FROM AUTHOR]

In this paper, a neural networks model for quantum computer is proposed. The core of this model is quantum neuron. Firstly, the inner product of the input qubits and the weight qubits is mapped to the phase of the control qubits in the neuron by the swap test technology, and then these phases are obtained by the phase estimation method, which are further used as the phase of the output qubit in the neuron. In this way, the mapping of input qubits to output qubit in quantum neuron is completed. The quantum neurons mentioned above can be used to construct quantum neural networks. In this paper, the quantum circuit for each operation step are given. The simulation results on the classic computer verify the effectiveness of the proposed model. [ABSTRACT FROM AUTHOR]

Abstract This paper addresses the issue of multistability for competitive neural networks. First, a general class of continuous non-monotonic piecewise linear activation functions is introduced. Then, based on the fixed point theorem, the contraction mapping theorem and the eigenvalue properties of strict diagonal dominance matrix, it is shown that under some conditions, such n -neuron competitive neural networks have exactly 5 n equilibrium points, among which 3 n equilibrium points are locally exponentially stable and the others are unstable. Moreover, it is revealed that the neural networks with non-monotonic piecewise linear activation functions introduced in this paper can have greater storage capacity than the ones with Mexican-hat-type activation function and nondecreasing saturated activation function. In addition, unlike most existing multistability results of neural networks with nondecreasing activation functions, the location of those obtained 3 n locally stable equilibrium points in this paper is more flexible. Finally, a numerical example is provided to illustrate and validate the theoretical findings via comprehensive computer simulations. [ABSTRACT FROM AUTHOR]

Abstract "Learning once, remembering forever", this wonderful cognitive phenomenon sometimes occurs in the learning process of human beings. Psychologists call this psychological phenomenon "one-trial learning". The traditional artificial neural networks can simulate the psychological phenomenon of "implicit learning", but can't simulate the cognitive phenomenon of "one-trial learning". Therefore, cognitive psychology gives a challenge to the traditional artificial neural networks. From two aspects of theory and practice in this paper, the possibility of simulating this kind of psychological phenomenon was explored by using morphological neural networks. This paper takes advantage of morphological associative memory networks to realize the simulation of "one-trial learning" for the first time, and gives 5 simulating practical examples. Theoretical analysis and simulation experiments show that the morphological associative memory networks are a higher effective machine learning method, and can better simulate the cognitive phenomenon of "one-trial learning", therefore provide a theoretical basis and technological support for the study of intelligent science and cognitive science. [ABSTRACT FROM AUTHOR]

Abstract In this paper, special memristor circuit and memristive retina network structure for analogue image processing have been presented. The new developments on memristor element have opened various possibilities due to its being in nano-scale and having nonlinear behavior. The proposed memristor emulator consists of only one Operational Transconductance Amplifier (OTA) and two MOS transistors which are operated in sub-threshold region. The memristor fuse structure is obtained by connecting two proposed memristor emulators. In the second section of our paper, we proposed a circuit block which is composed of 16 × 16 pixels retinomorphic memristive grid to maintain a smoothed and edges preserved image. All simulation results for both proposed memristor circuits and retinomorphic grid are obtained as expected. [ABSTRACT FROM AUTHOR]

Abstract Data Mining (DM) is a fundamental component of the Data Science process. Over recent years a huge library of DM algorithms has been developed to tackle a variety of problems in fields such as medical imaging and traffic analysis. Many DM techniques are far more flexible than more classical numerial simulation or statistical modelling approaches. These could be usefully applied to data-rich environmental problems. Certain techniques such as artificial neural networks, clustering, case-based reasoning or Bayesian networks have been applied in environmental modelling, while other methods, like support vector machines among others, have yet to be taken up on a wide scale. There is greater scope for many lesser known techniques to be applied in environmental research, with the potential to contribute to addressing some of the current open environmental challenges. However, selecting the best DM technique for a given environmental problem is not a simple decision, and there is a lack of guidelines and criteria that helps the data scientist and environmental scientists to ensure effective knowledge extraction from data. This paper provides a broad introduction to the use of DM in Data Science processes for environmental researchers. Data Science contains three main steps (pre-processing, data mining and post-processing). This paper provides a conceptualization of Environmental Systems and a conceptualization of DM methods, which are in the core step of the Data Science process. These two elements define a conceptual framework that is on the basis of a new methodology proposed for relating the characteristics of a given environmental problem with a family of Data Mining methods. The paper provides a general overview and guidelines of DM techniques to a non-expert user, who can decide with this support which is the more suitable technique to solve their problem at hand. The decision is related to the bidimensional relationship between the type of environmental system and the type of DM method. An illustrative two way table containing references for each pair Environmental System-Data Mining method is presented and discussed. Some examples of how the proposed methodology is used to support DM method selection are also presented, and challenges and future trends are identified. Highlights • A methodology to support Data Mining (DM) method choice in environmental sciences. • DMMCM is a conceptual map providing an overview of common Data Mining methods. • DMMT provides templates of main DM families, to simplify use by practicioners. • Extensive real environmental applications and 5 detailed case studies are shown. • Existing and future challenges in DM for Data Science are postulated. [ABSTRACT FROM AUTHOR]

To fully understand the possibilities and the limits of the Circular Economy (CE), a comprehensive model taking into account its different stages (product design, mechanical pre-processing, metallurgy, etc.) is required. A crucial aspect is to understand the inevitable losses at different stages of recycling. The complexity of the material streams in mechanical separation processes requires a detailed description of particles and their properties to successfully simulate unit processes. This paper presents a new approach that connects measurement-based particle properties to statistical modelling and simulation of mechanical separation processes. The proposed approach combines particle tracking with the generalization ability of neural networks. Above all, it advances the present particle binning and tracking methods utilizing property-based binning rather than liberation-based binning for modelling purposes of complex systems. In order to demonstrate the new approach, this paper uses Mineral Liberation Analysis (MLA) data from magnetic and gravity separation processes of a complex ore. The proposed approach can be integrated into present simulation platforms such as HSC Sim. [ABSTRACT FROM AUTHOR]

In this paper, without separating the complex-valued neural networks into two real-valued systems, the quasi-projective synchronization of fractional-order complex-valued neural networks is investigated. First, two new fractional-order inequalities are established by using the theory of complex functions, Laplace transform and Mittag-Leffler functions, which generalize traditional inequalities with the first-order derivative in the real domain. Additionally, different from hybrid control schemes given in the previous work concerning the projective synchronization, a simple and linear control strategy is designed in this paper and several criteria are derived to ensure quasi-projective synchronization of the complex-valued neural networks with fractional-order based on the established fractional-order inequalities and the theory of complex functions. Moreover, the error bounds of quasi-projective synchronization are estimated. Especially, some conditions are also presented for the Mittag-Leffler synchronization of the addressed neural networks. Finally, some numerical examples with simulations are provided to show the effectiveness of the derived theoretical results. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) is an extensive scientific discipline which enables computer systems to solve problems by emulating complex biological processes such as learning, reasoning and self-correction. This paper presents a comprehensive review of the application of AI techniques for improving performance of optical communication systems and networks. The use of AI-based techniques is first studied in applications related to optical transmission, ranging from the characterization and operation of network components to performance monitoring, mitigation of nonlinearities, and quality of transmission estimation. Then, applications related to optical network control and management are also reviewed, including topics like optical network planning and operation in both transport and access networks. Finally, the paper also presents a summary of opportunities and challenges in optical networking where AI is expected to play a key role in the near future. [ABSTRACT FROM AUTHOR]

This paper deals with a class of neutral shunting inhibitory cellular neural networks with time-varying delays and D operator. Using the differential inequality theory and Lyapunov functional method, a set of sufficient conditions which ascertains that the existence and exponential stability of anti-periodic solutions of neutral shunting inhibitory cellular neural networks with time-varying delays and D operator are derived. Computer simulations are delineated to substantiate the correctness of our theoretical predictions. The obtained results of this paper are new and complement some earlier works. [ABSTRACT FROM AUTHOR]

Some of the authors of this paper had already established a pedal operated human powered flywheel motor (HPFM) which justifies the energy requirements for process units. The different types of process units designed and tested so far intestinally suits to rural areas such as brick making machine, Low head water lifting, Wood turning, Wood strips cutting, electricity generation etc. This machine system includes three sub systems namely (i) HPFM (ii) Torsionally Flexible Clutch (TFC) (iii) A Process Unit. ANN modeling has been used to model the experimental findings for human powered flywheel motor. It has been observed that neuron size, transfer function, training function plays important role in performance of the network. The optimal selection of parametric values of each ANN parameter is carried through observation of performance, regression plots. This paper illustrates a unique method of selecting optimal ANN network configuration for fitting function approximation problem. We also found that reliability of the derived ANN model is 97%. [ABSTRACT FROM AUTHOR]

The subsea capping stack is crucial equipment for emergency rescue in a blowout scenario, which determines the success or failure of the emergency rescue. The secondary risk prevention of the blowout emergency continues to face a persistent challenge. This paper proposes a methodology framework for emergency risk of the subsea capping stack, which aims to realize the quantitative risk assessment of installation operation by integrating numerical simulation and an Artificial neural network (ANN) model. Herein, the numerical model of fluid impact for the capping stack is constructed, and the impact forces of the blowout fluid are analyzed. The calculation results indicate that the blowout pressure, water depth, and installation height significantly affect the impact force during emergency operations. Furthermore, a genetic algorithm optimizing the neural network (GA-BP) model for quickly predicting the blowout impact force is established to improve emergency response efficiency based on sufficient simulation data. The prediction model has a minor mean absolute percentage error (MAPE), and its accuracy is also verified compared with the SVM model. Besides, the developed model speeds up the processing time from hours to seconds compared to numerical simulation. Therefore, it can provide a decision-making basis for emergency engineers at the site of blowout accidents more efficiently. • An integrated methodology is proposed to quantify the installation risk for the subsea capping stack. • The simulation model of the capping stack during the installation process is constructed to predict the impact forces. • A prediction model is developed to speed up the processing time from hours to seconds compared to numerical simulation. [ABSTRACT FROM AUTHOR]

In this paper, without transforming the second order inertial neural networks into the first order differential systems by some variable substitutions, asymptotic stability and synchronization for a class of delayed inertial neural networks are investigated. Firstly, a new Lyapunov functional is constructed to directly propose the asymptotic stability of the inertial neural networks, and some new stability criteria are derived by means of Barbalat Lemma. Additionally, by designing a new feedback control strategy, the asymptotic synchronization of the addressed inertial networks is studied and some effective conditions are obtained. To reduce the control cost, an adaptive control scheme is designed to realize the asymptotic synchronization. It is noted that the dynamical behaviors of inertial neural networks are directly analyzed in this paper by constructing some new Lyapunov functionals, this is totally different from the traditional reduced-order variable substitution method. Finally, some numerical simulations are given to demonstrate the effectiveness of the derived theoretical results. [ABSTRACT FROM AUTHOR]

In this paper, the synchronization of memristor-based neural networks with time-varying delays via pinning control is investigated. A novel pinning method is introduced to synchronize two memristor-based neural networks which denote drive system and response system, respectively. The dynamics are studied by theories of differential inclusions and nonsmooth analysis. In addition, some sufficient conditions are derived to guarantee asymptotic synchronization and exponential synchronization of memristor-based neural networks via the presented pinning control. Furthermore, some improvements about the proposed control method are also discussed in this paper. Finally, the effectiveness of the obtained results is demonstrated by numerical simulations. [ABSTRACT FROM AUTHOR]

In this paper, two new Artificial Neural Network MPPT controllers based on fixed and variable step size have been proposed and investigated. The data required to generate the ANN model are generated using the classical Perturbation and Observation algorithm. The neural network MPPT controller is developed in two steps: the offline step required for training of different neural networks parameters in order to find the optimal neural network MPPT controller (structure, activation function and training algorithm) and the Online step where the optimal neural network MPPT controller is used in PV system. The performance of the proposed variable step size and fixed step size ANN-MPPT methods are analyzed under different operating conditions using Matlab/Simulink. To validate the simulated system hardware implementation of the proposed algorithms was carried out using experimental prototype MPPT based on Flyback converter connected to Solarex MSX-60 (4 panels) and dsPIC30F4011 control circuit. Analysis and comparative study between the proposed fixed and variable step size ANN-MPPT controllers have been presented, showing a real contributions in term of tracking accuracy, response time, overshoot and steady state ripple. In addition, this paper can be considered as a review study on ANN-MPPT methods for PV systems. [ABSTRACT FROM AUTHOR]

In this paper, the coexistence and dynamical behaviors of multiple equilibrium points are discussed for a class of memristive neural networks (MNNs) with unbounded time-varying delays and nonmonotonic piecewise linear activation functions. By means of the fixed point theorem, nonsmooth analysis theory and rigorous mathematical analysis, it is proven that under some conditions, such n -neuron MNNs can have 5 n equilibrium points located in ℜ n , and 3 n of them are locally μ -stable. As a direct application, some criteria are also obtained on the multiple exponential stability, multiple power stability, multiple log-stability and multiple log–log-stability. All these results reveal that the addressed neural networks with activation functions introduced in this paper can generate greater storage capacity than the ones with Mexican-hat-type activation function. Numerical simulations are presented to substantiate the theoretical results. [ABSTRACT FROM AUTHOR]

This paper presents a comprehensive overview of modelling, simulation and implementation of neural networks, taking into account that two aims have emerged in this area: the improvement of our understanding of the behaviour of the nervous system and the need to find inspiration from it to build systems with the advantages provided by nature to perform certain relevant tasks. The development and evolution of different topics related to neural networks is described (simulators, implementations, and real-world applications) showing that the field has acquired maturity and consolidation, proven by its competitiveness in solving real-world problems. The paper also shows how, over time, artificial neural networks have contributed to fundamental concepts at the birth and development of other disciplines such as Computational Neuroscience, Neuro-engineering, Computational Intelligence and Machine Learning. A better understanding of the human brain is considered one of the challenges of this century, and to achieve it, as this paper goes on to describe, several important national and multinational projects and initiatives are marking the way to follow in neural-network research. [ABSTRACT FROM AUTHOR]

Using accurate models of photovoltaic modules is of major importance to make realistic simulations of these systems in order to study their elements for a better performance. In this paper we address the problem of the lack of a systematic procedure to obtain accurate artificial neural network based models in an unattended way with large datasets. We face this problem introducing a novel systematic procedure to carry out this task. As proof of concept, we tested the procedure modeling a Mitsubishi Electric PV-TD185MF5 (185 Wp) photovoltaic module placed at the University College of Engineering of Vitoria-Gasteiz (University of the Basque Country, Spain). We have used a dataset of 63,000 samples collected during 18 months (from August 2013 to February 2015). The main findings of the paper are two. The first one is that the systematic procedure works properly because it generated autonomously two very accurate models of I PV (RMSE with unknown test dataset of 0.045 A for a one hidden layer neural network, while 0.042 A for a two hidden layers neural network), i.e., we conclude that the unattended execution of the systematic procedure introduced in this paper has obtained models which have learned the electrical behavior of the photovoltaic module with an accuracy higher than the measurement devices precision. We have compared these results with recent relevant papers and we found that the proposed procedure is competitive and improves the state-of-art results. The second finding is that using these models lead to space savings larger than 99.5% of the original tabular representation of the dataset. [ABSTRACT FROM AUTHOR]

Maintaining and enhancing engagement of patients during stroke rehabilitation exercises are in the focus of current research. There have been various methods and computer supported tools developed for this purpose, which try to avoid mundane exercising that is prone to become a routine or even boring for the patients and leads to ineffective training. This paper proposes a strategy bundle-based smart learning mechanism (SLM) to increase the efficiency of rehabilitation exercises. The underpinning strategy considers motor, perceptive, cognitive and emotional aspects of engagement. Part of a cyber-physical stroke rehabilitation system (CP-SRS), the proposed SLM is able to learn the relationship between the actual engagement levels and applied stimulations. From a computational point of view, the SLM is based on multiplexed signal processing and a machine learning agent. The paper presents the mathematical concepts of signal processing, the reasoning algorithms, and the overall embedding of the SLM in the CP-SRS. Regression and classification are two possible solutions for this learning mechanism. Computer simulation is conducted to investigate the limitations of the proposed learning mechanism and compare the results of different machine learning methods. We simulate regression with artificial neural network (ANN), and classification with ANN and Naive Bayes (NB). Results show that classification with NB is more promising in practice since it is less sensitive to the deviations in the inputs than the applied version of ANN. [ABSTRACT FROM AUTHOR]

Sequential processes appear naturally in all types of industries. Biotechnology is a good example of such schemes. Wastewater treatment using microbiological activity is a particular case having all the characteristics of sequential methods. Sulfate reduction as pre-treatment followed by the decomposition of sulfated compounds using adapted microorganisms is the sequential nonlinear process with state constrains analyzed in this paper. Modeling this procedure is still a difficult task because the number of elements involved in the reaction. This paper presents an adaptive algorithm to obtain a suitable model of this process using continuous neural networks. The adaptive model preserves the sequential nature of the process as well as the bounded nature of all states. The neural network is proposed as a system identifier in terms of the hybrid systems theory. The Lyapunov stability method is used to demonstrate the convergence of the identifier states to the real concentrations of the microbiological system. Experimental results and their corresponding simulation using the adaptive model based on neural networks confirm the theoretical results described in this paper. [ABSTRACT FROM AUTHOR]

The problem of exponential stabilization and synchronization for fuzzy model of memristive neural networks (MNNs) is investigated by using periodically intermittent control in this paper. Based on the knowledge of memristor and recurrent neural network, the model of MNNs is formulated. Some novel and useful stabilization criteria and synchronization conditions are then derived by using the Lyapunov functional and differential inequality techniques. It is worth noting that the methods used in this paper are also applied to fuzzy model for complex networks and general neural networks. Numerical simulations are also provided to verify the effectiveness of theoretical results. [ABSTRACT FROM AUTHOR]

This paper is concerned with the periodic synchronization problem for a general class of delayed neural networks (DNNs) with discontinuous neuron activation. One of the purposes is to analyze the problem of periodic orbits. To do so, we introduce new tools including inequality techniques and Kakutani’s fixed point theorem of set-valued maps to derive the existence of periodic solution. Another purpose is to design a switching state-feedback control for realizing global exponential synchronization of the drive–response network system with periodic coefficients. Unlike the previous works on periodic synchronization of neural network, both the neuron activations and controllers in this paper are allowed to be discontinuous. Moreover, owing to the occurrence of delays in neuron signal, the neural network model is described by the functional differential equation. So we introduce extended Filippov-framework to deal with the basic issues of solutions for discontinuous DNNs. Finally, two examples and simulation experiments are given to illustrate the proposed method and main results which have an important instructional significance in the design of periodic synchronized DNNs circuits involving discontinuous or switching factors. [ABSTRACT FROM AUTHOR]

Construction sites are changing every day, which brings some difficulties for different contractors to do their tasks properly. One of the key points for all entities who work on the same site is the location of resources including materials, tools, and equipment. Therefore, the lack of an integrated localization system leads to an increase in the time wasted on searching for resources. In this research, a localization method based on Radio Frequency Identification (RFID) systems which does not need infrastructure is proposed to overcome this problem. This paper investigates the usage of active RFID technology for the localization of movable objects (e.g. components, equipment, and tools) equipped with RFID tags using handheld readers by extending a Cluster-based Movable Tag Localization (CMTL) technique which uses a k-Nearest Neighbor (k-NN) algorithm. CMTL uses a multidimensional clustering technique that considers signal pattern similarity between target and reference tags together with spatial distribution of reference tags for detecting the region where the target tag is located. The paper proposes applying an irregular bilinear interpolation method to form a grid of virtual reference tags within the selected cluster of real reference tags. Moreover, the proposed method uses artificial neural networks (ANNs) for positioning the target tag, as opposed to empirical weighted averaging formulas used in similar k-NN based methods. Comparative analysis is performed to quantify the improvement of the proposed method over similar k-NN-based methods using a simulation environment. A case study is performed to analyze the performance of the proposed method. [ABSTRACT FROM AUTHOR]

This paper is concerned with the problem of coexistence and dynamical behaviors of multiple equilibrium points for neural networks with discontinuous non-monotonic piecewise linear activation functions and time-varying delays. The fixed point theorem and other analytical tools are used to develop certain sufficient conditions that ensure that the n -dimensional discontinuous neural networks with time-varying delays can have at least 5 n equilibrium points, 3 n of which are locally stable and the others are unstable. The importance of the derived results is that it reveals that the discontinuous neural networks can have greater storage capacity than the continuous ones. Moreover, different from the existing results on multistability of neural networks with discontinuous activation functions, the 3 n locally stable equilibrium points obtained in this paper are located in not only saturated regions, but also unsaturated regions, due to the non-monotonic structure of discontinuous activation functions. A numerical simulation study is conducted to illustrate and support the derived theoretical results. [ABSTRACT FROM AUTHOR]

In this paper existence and asymptotic stability of asymptotically almost automorphic solution of impulsive neural networks with delay is discussed. The results are established by using various fixed point theorems and Lyapunov-like functional. As far as we know, this is the first paper to discuss such kind of solutions for impulsive neural networks. At the end, we give few examples to illustrate our theoretical findings. One can see that the numerical simulation results show asymptotically almost automorphic behaviour of the solution. [ABSTRACT FROM AUTHOR]

Neural network exhibits superior ability in mapping and finding input-output patterns from a priori data set. Neural network is a common data analytics tool for extracting knowledge and forecasting. In this paper, a General Regression Neural Network or GRNN is used in modeling and forecasting Global Innovation Output. Global Innovation Index (GII) provides a detailed analysis of underlying factors influencing year-on-year changes in rankings; identifies the strengths and weaknesses of each country's profile; computes and provides the national indicators based on the certain input and output indicators. This paper particularly focuses on Innovation Output of United Arab Emirates or UAE. General Regression Neural Networks is proposed using GII data for 125 countries for training and then optimized to compute the improvement on the innovation output indicator of UAE as a test case by individually and collectively using following inputs from Singapore: Regulatory environment, Tertiary education, Information and Communication Technologies, Market sophistication in Investment and Knowledge absorption. It is observed that the output indicator improves from a value of 29.6 to about 42 when Knowledge absorption input is considered individually whereas the output increases to over 45 when all five indicators are simultaneously improved. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, a finite-time lag synchronization of coupled neural networks with time delay is investigated. By means of the Lyapunov stability theory, a feedback controller is designed for achieving lag synchronization between two delayed neural networks systems in finite time. Paper extends some basic results from the area of finite time to time-delay systems. Numerical simulations on coupled Lu neural systems illustrate the effectiveness of the results. [Copyright &y& Elsevier]

Abstract This paper investigates the coexistence and local stability of multiple equilibrium points for competitive neural networks, where the Gaussian-wavelet-type activation functions are employed and the unbounded time-varying delays are considered. Based on geometric formulation, the fixed point theorem, contraction mapping theorem and rigorous mathematical analysis, a series of sufficient conditions are derived to ascertain that the addressed neural networks have exactly 5 n equilibrium points, among which 3 n equilibrium points are locally stable. On this basis, some criteria are also obtained on the multiple exponential stability, multiple power stability and multiple log-stability of Hopfield neural networks with Gaussian-wavelet-type activation functions. The obtained results generalize and improve the existing multistability results of Hopfield neural networks and competitive neural networks without time delays and with Gaussian-wavelet-type activation functions. Moreover, it is highlighted that the competitive neural networks with Gaussian-wavelet-type activation functions can have both more total equilibrium points and more locally stable equilibrium points than the ones with Mexican-hat-type activation function. Finally, two numerical examples with computer simulations are provided to illustrate and validate the theoretical results. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, a special class of recurrent neural network, termed Zhang neural network (ZNN), is investigated for the online solution of the time-varying matrix pseudoinverse. Meanwhile, a novel activation function, named Li activation function, is employed. Then, based on two basic Zhang functions (ZFs) and the intrinsically nonlinear method of ZNN design, two finite-time convergent ZNN models (termed ZNN-1 model and ZNN-2 model) are first proposed and investigated for time-varying matrix pseudoinversion. Such two ZNN models can be accelerated to finite-time convergence to the time-varying theoretical pseudoinverse. The upper bound of the convergence time is also derived analytically via Lyapunov theory. By exploiting the other three simplified ZFs and the extended nonlinearization method, three simplified finite-time convergent ZNN models (termed ZNN-3 model, ZNN-4 model and ZNN-5 model) are sequentially proposed. In addition, the link between the ZNN models and the Getz–Marsden (G–M) dynamic system is discovered and presented in this paper. Computer-simulation results further substantiate the theoretical analysis and demonstrate the effectiveness of ZNN models based on different ZFs for the time-varying matrix pseudoinverse. [Copyright &y& Elsevier]

Abstract: In this paper, a one-layer recurrent projection neural network is proposed for solving pseudoconvex optimization problems with general convex constraints. The proposed network in this paper deals with the constraints into two parts, which brings the network simpler structure and better properties. By the Tikhonov-like regularization method, the proposed network need not estimate the exact penalty parameter in advance. Moreover, comparing with some existing neural networks, the proposed network can solve more general constrained pseudoconvex optimization problems. When the solution of the proposed network is bounded, it converges to the optimal solution set of considered optimization problem, which may be nonsmooth and nonconvex. Meantime, some sufficient conditions are presented to guarantee the boundedness of the solution of the proposed network. Numerical examples with simulation results are given to illustrate the effectiveness and good characteristics of the proposed network for solving constrained pseudoconvex optimization. [Copyright &y& Elsevier]

Abstract: This paper presents a global Gaussian mixture (GM) reduction algorithm via clustering for extended target tracking in clutter. The proposed global clustering algorithm is obtained by combining a fuzzy Adaptive Resonance Theory (ART) neural network architecture with the weighted Kullback–Leibler (KL) difference which describes discrimination of one component from another. Therefore, we call the proposed algorithm as ART-KL clustering (ART-KL-C) in the paper. The weighted KL difference is used as a category choice function of ART-KL-C, derived by considering both the KL divergence between two components and their weights. The performance of ART-KL-C is evaluated by the normalized integrated squared distance (NISD) measure, which describes the deviation between the original and reduced GM. The proposed algorithm is tested on both one-dimensional and four-dimensional simulation examples, and the results show that the proposed algorithm can more accurately approximate the original mixture and is useful in extended target tracking. [Copyright &y& Elsevier]

Abstract: This paper aims to develop a new idea for the classification of transient disturbances on power quality. The method is based on the combination of spectral kurtosis (SK) and artificial neural network (ANN). The SK is a high-order statistical moment which can detect the non-Gaussian components in a signal. Through the introduction of SK and its properties, we propose a classification plan for five transient disturbances combining SK and ANN. Firstly, the high frequency parts of five disturbance signals are extracted with DB4 wavelet transform (WT). Secondly, their SK values are respectively computed based on short time Fourier transform (STFT) and WT. Because the features of SK based on WT for five disturbance signals are not clearly distinguished, we propose a new computation method of SK based on Butterworth Distribution (BUD). Lastly, we choose the maximum, minimum and average values of SK based on STFT and BUD as the eigenvectors for the transient disturbance classification, which are input into RBF neural network. The simulation results show that the recognition rate of five transient disturbances is high, and the classification method proposed in the paper for transient power quality combining SK with ANN is efficient and feasible. [Copyright &y& Elsevier]

Abstract: In this paper, we investigate the global dynamics of equilibrium point for delayed competitive neural networks with different time scales and discontinuous activations. Employing the Leray–Schauder alternative theorem in multivalued analysis, linear matrix inequality technique and generalized Lyapunov-like method, we obtain some new sufficient conditions ensuring the existence, uniqueness and global stability and the global convergence in finite time of the equilibrium point, the results of this paper improve and extend previously known results. Finally, two examples and numerical simulations are conducted to validate the theoretical results. [Copyright &y& Elsevier]

Abstract: The paper studies several related problems of the Extended Alternating Projection Neural Network (EAPNN). Firstly the mathematical relation between the interconnection matrix and the library pattern matrix, and the stability of the EAPNN with infinite precision data have been studied, strictly mathematical proofs for them have been given. After analyzing the finite precision influence on the training method of the EAPNN, the paper proposes an improved training method for the EAPNN with finite precision data and gives a practical threshold-value determination method. Then the mathematical relation among the factors influencing the association times has been studied and the mathematical expression of the association times for reaching the expected association precision, while network being used for CAM, has been deduced. Finally simulation experiments have been designed and simulation results demonstrate the validity of theoretical analyses. [Copyright &y& Elsevier]

Abstract: The objective of this paper is to present a novel method to achieve maximum reliability for fault tolerant optimal network design when network has variable size. Reliability calculation is most important and critical component when fault tolerant optimal network design is required. A network must be supplied with certain parameters that guarantee proper functionality and maintainability under worse situations. Many alternative methods for measuring reliability have been stated in literature for optimal network design. Most of these methods mentioned in literature for evaluating reliability may be analytical and simulation based. These methods provide significant way to compute reliability when network has limited size. Also, significant computational effort is required for growing variable sized networks. Therefore, a novel neural network method is presented to achieve significant high reliability for fault tolerant optimal network design in highly growing variable networks. This paper computes reliability with improved learning rate gradient descent based neural network method. The result shows that improved optimal network design with maximum reliability is achievable by novel neural network at manageable computational cost. [Copyright &y& Elsevier]

Abstract: On-line learning and recognition of spatio- and spectro-temporal data (SSTD) is a very challenging task and an important one for the future development of autonomous machine learning systems with broad applications. Models based on spiking neural networks (SNN) have already proved their potential in capturing spatial and temporal data. One class of them, the evolving SNN (eSNN), uses a one-pass rank-order learning mechanism and a strategy to evolve a new spiking neuron and new connections to learn new patterns from incoming data. So far these networks have been mainly used for fast image and speech frame-based recognition. Alternative spike-time learning methods, such as Spike-Timing Dependent Plasticity (STDP) and its variant Spike Driven Synaptic Plasticity (SDSP), can also be used to learn spatio-temporal representations, but they usually require many iterations in an unsupervised or semi-supervised mode of learning. This paper introduces a new class of eSNN, dynamic eSNN, that utilise both rank-order learning and dynamic synapses to learn SSTD in a fast, on-line mode. The paper also introduces a new model called deSNN, that utilises rank-order learning and SDSP spike-time learning in unsupervised, supervised, or semi-supervised modes. The SDSP learning is used to evolve dynamically the network changing connection weights that capture spatio-temporal spike data clusters both during training and during recall. The new deSNN model is first illustrated on simple examples and then applied on two case study applications: (1) moving object recognition using address-event representation (AER) with data collected using a silicon retina device; (2) EEG SSTD recognition for brain–computer interfaces. The deSNN models resulted in a superior performance in terms of accuracy and speed when compared with other SNN models that use either rank-order or STDP learning. The reason is that the deSNN makes use of both the information contained in the order of the first input spikes (which information is explicitly present in input data streams and would be crucial to consider in some tasks) and of the information contained in the timing of the following spikes that is learned by the dynamic synapses as a whole spatio-temporal pattern. [Copyright &y& Elsevier]

Abstract: In this paper we focus on modeling autonomous learning to improve performance of a humanoid robot through a modular artificial neural networks architecture. A model of a neural controller is presented, which allows a humanoid robot iCub to autonomously improve its sensorimotor skills. This is achieved by endowing the neural controller with a secondary neural system that, by exploiting the sensorimotor skills already acquired by the robot, is able to generate additional imaginary examples that can be used by the controller itself to improve the performance through a simulated mental training. Results and analysis presented in the paper provide evidence of the viability of the approach proposed and help to clarify the rational behind the chosen model and its implementation. [Copyright &y& Elsevier]

Abstract: In this paper, reaction–diffusion Hopfield neural networks systems with time-varying delays and Dirichlet boundary conditions are investigated. The theorems on existence and global exponential stability of anti-periodic mild solution are established. Moreover, theoretical results further show that diffusion terms contribute to existence and stabilization of anti-periodic mild solution. Finally, an illustrative example and numerical simulations are given to show effectiveness of results in this paper. [Copyright &y& Elsevier]

Abstract: This paper proposes a system for wind turbine condition monitoring using Adaptive Neuro-Fuzzy Interference Systems (ANFIS). For this purpose: (1) ANFIS normal behavior models for common Supervisory Control And Data Acquisition (SCADA) data are developed in order to detect abnormal behavior of the captured signals and indicate component malfunctions or faults using the prediction error. 33 different standard SCADA signals are used and described, for which 45 normal behavior models are developed. The performance of these models is evaluated in terms of the prediction error standard deviations to show the applicability of ANFIS models for monitoring wind turbine SCADA signals. The computational time needed for model training is compared to Neural Network (NN) models showing the strength of ANFIS in training speed. (2) For automation of fault diagnosis Fuzzy Interference Systems (FIS) are used to analyze the prediction errors for fault patterns. The outputs are both the condition of the component and a possible root cause for the anomaly. The output is generated by the aid of rules that capture the existing expert knowledge linking observed prediction error patterns to specific faults. The work is based on continuously measured wind turbine SCADA data from 18 turbines of the 2 MW class covering a period of 30 months. The system proposed in this paper shows a novelty approach with regard to the usage of ANFIS models in this context and the application of the proposed procedure to a wide range of SCADA signals. The applicability of the set up ANFIS models for anomaly detection is proved by the achieved performance of the models. In combination with the FIS the prediction errors can provide information about the condition of the monitored components. In this paper the condition monitoring system is described. Part two will entirely focus on application examples and further efficiency evaluation of the system. [Copyright &y& Elsevier]

Abstract: The market for small Solar CombiSystems (SCSs) is penalized by its lack of common performance test method. The methodology proposed in this paper is based on the SCSPT procedure which tests each combisystem as a whole system on a semi-virtual test bench. Test results are used to identify a “Gray Box” model of the tested SCS, which includes an Artificial Neural Network. This model can then simulate the behavior of the system for any climate and any building. Results of all those simulations are finally used according to the FSC procedure in order to characterize the tested SCS performances with a simple curve. An experimental study of this methodology with two real SCS is presented in this paper. [Copyright &y& Elsevier]

Abstract: The paper analyzes a general memristor-based recurrent neural networks with time-varying delays (DRNNs). The dynamic analysis in the paper employs results from the theory of differential equations with discontinuous right-hand side as introduced by Filippov, and some new conditions concerning global exponential stability are obtained. In addition, these conditions do not require the activation functions to be differentiable, the connection weight matrices to be symmetric and the delay functions to be differentiable, our results are mild and more general. Finally, numerical simulations illustrate the effectiveness of our results. [Copyright &y& Elsevier]

Abstract: Multiple Sensor fusion is important for human and animals to localize stimuli within an environment. Visual and audio sensors are often integrated to direct attention such as in human communication. The paper describes a methodology for designing and developing architecture to effectively localise simultaneous audio and visual stimuli by integrating both sensors. The Superior Colliculus (SC) inspired the architecture and it mimics the top and deep layers of the SC, as these layers are mainly responsible for visual and audio stimuli localization, respectively. The integration methodology described in this paper is evaluated against algorithmic-based methodology to determine effectiveness of the approach. Experimental results successfully demonstrate the key advantages in the integration, including (i) low-level multimodal stimuli localization and (ii) dimensionality reduction of the input-space without affecting stimuli strength and localization. [Copyright &y& Elsevier]

Abstract: Distal reward refers to a class of problems where reward is temporally distal from actions that lead to reward. The difficulty for any biological neural system is that the neural activations that caused an agent to achieve reward may no longer be present when the reward is experienced. Therefore in addition to the usual reward assignment problem, there is the additional complexity of rewarding through time based on neural activations that may no longer be present. Although this problem has been thoroughly studied over the years using methods such as reinforcement learning, we are interested in a more biologically motivated neural architectural approach. This paper introduces one such architecture that exhibits rudimentary distal reward learning based on associations of bottom-up visual sensory sequences with bottom-up proprioceptive motor sequences while an agent explores an environment. After sufficient learning, the agent is able to locate the reward through chaining together of top-down motor command sequences. This paper will briefly discuss the details of the neural architecture, the agent-based modeling system in which it is embodied, a virtual Morris water maze environment used for training and evaluation, and a sampling of numerical experiments characterizing its learning properties. [Copyright &y& Elsevier]

Abstract: This paper is concerned with the stability of delayed recurrent neural networks with impulse control and Markovian jump parameters. The jumping parameters are modeled as a continuous-time, discrete-state Markov process. By applying the Lyapunov stability theory, Dynkin’s formula and linear matrix inequality technique, some new delay-dependent conditions are derived to guarantee the exponential stability of the equilibrium point. Moreover, three numerical examples and their simulations are given to show the less conservatism and effectiveness of the obtained results. In particular, the traditional assumptions on the differentiability of the time varying delays and the boundedness of their derivatives are removed since the time varying delays considered in this paper may not be differentiable, even not continuous. [Copyright &y& Elsevier]

Abstract: This paper investigates the batch back-propagation algorithm with penalty for training feedforward neural networks. A usual penalty is considered, which is a term proportional to the norm of the weights. The learning rate is set to be a small constant or an adaptive series. The main contribution of this paper is to theoretically prove the boundedness of the weights in the network training process. This boundedness is then used to prove some convergence results of the algorithm, which cover both the weak and strong convergence. Simulation results are given to support the theoretical findings. [Copyright &y& Elsevier]