Search

This paper presents an improved method for reducing high-order dynamical system models via clustering. Agglomerative hierarchical clustering based on performance evaluation (HC-PE) is introduced for model order reduction. This method computes the reduced order denominator of the transfer function model by clustering system poles in a hierarchical dendrogram. The base layer represents an $n{\textrm {th}}$ order system, which is used to calculate each successive layer to reduce the model order until finally reaching a second-order system. HC-PE uses a mean-squared error (MSE) in every reduced order, which modifies the pole placement process. The coefficients for the numerator of the reduced model are calculated by using the Padé approximation (PA) or alternatively a genetic algorithm (GA). Several numerical examples of reducing techniques are taken from the literature to compare with HC-PE. Two classes of results are shown in this paper. The first sets are single-input single-output models that range from simple models to 48th order systems. The second sets of experiments are with a multi-input multioutput model. We demonstrate the best performance for HC-PE through minimum MSEs compared with other methods. Furthermore, the robustness of HC-PE combined with PA or GA is confirmed by evaluating the third-order reduced model for the triple-link inverted pendulum model by adding a disturbance impulse signal and by changing model parameters. The relevant stability proofs are provided in Appendixes A and B in the supplementary material. HC-PE with PA slightly outperforms its performance with GA, but both approaches are attractive alternatives to other published methods. [ABSTRACT FROM AUTHOR]

It is a widely accepted fact that the computational capability of recurrent neural networks (RNNs) is maximized on the so-called “edge of criticality.” Once the network operates in this configuration, it performs efficiently on a specific application both in terms of: 1) low prediction error and 2) high short-term memory capacity. Since the behavior of recurrent networks is strongly influenced by the particular input signal driving the dynamics, a universal, application-independent method for determining the edge of criticality is still missing. In this paper, we aim at addressing this issue by proposing a theoretically motivated, unsupervised method based on Fisher information for determining the edge of criticality in RNNs. It is proved that Fisher information is maximized for (finite-size) systems operating in such critical regions. However, Fisher information is notoriously difficult to compute and requires the analytic form of the probability density function ruling the system behavior. This paper takes advantage of a recently developed nonparametric estimator of the Fisher information matrix and provides a method to determine the critical region of echo state networks (ESNs), a particular class of recurrent networks. The considered control parameters, which indirectly affect the ESN performance, are explored to identify those configurations lying on the edge of criticality and, as such, maximizing Fisher information and computational performance. Experimental results on benchmarks and real-world data demonstrate the effectiveness of the proposed method. [ABSTRACT FROM PUBLISHER]

Traffic congestion often incurs environmental problems. One of the most effective ways to mitigate this is carpooling transportation, which substantially reduces automobile demands. Due to the popularization of smartphones and mobile applications, a carpool service can be conveniently accessed via the intelligent carpool system. In this system, the service optimization required to intelligently and adaptively distribute the carpool participant resources is called the carpool service problem (CSP). Several previous studies have examined viable and preliminary solutions to the CSP by using exact and metaheuristic optimization approaches. For CSP-solving, evolutionary computation (e.g., metaheuristics) is a more promising option in comparison to exact-type approaches. However, all the previous state-of-the-art approaches use pure optimization to solve the CSP. In this paper, we employ the framework of neuroevolution to propose the self-organizing map-based neuroevolution (SOMNE) solver by which the SOM-like network represents the abstract CSP solution and is well-trained by using neural learning and evolutionary mechanism. The experimental section of this paper investigates the comparisons and analyses of two objective functions of the CSP and demonstrates that the proposed SOMNE solver achieves superior results when compared against those the other approaches produce, especially in regard to the optimization of the primary objective functions of the CSP. Finally, the visual results of the SOM are illustrated to show the effectiveness and efficiency of the evolutionary neural learning process. [ABSTRACT FROM AUTHOR]

Conventional feature selection algorithms assign a single common feature set to all regions of the sample space. In contrast, this paper proposes a novel algorithm for localized feature selection for which each region of the sample space is characterized by its individual distinct feature subset that may vary in size and membership. This approach can therefore select an optimal feature subset that adapts to local variations of the sample space, and hence offer the potential for improved performance. Feature subsets are computed by choosing an optimal coordinate space so that, within a localized region, within-class distances and between-class distances are, respectively, minimized and maximized. Distances are measured using a logistic function metric within the corresponding region. This enables the optimization process to focus on a localized region within the sample space. A local classification approach is utilized for measuring the similarity of a new input data point to each class. The proposed logistic localized feature selection (lLFS) algorithm is invariant to the underlying probability distribution of the data; hence, it is appropriate when the data are distributed on a nonlinear or disjoint manifold. lLFS is efficiently formulated as a joint convex/increasing quasi-convex optimization problem with a unique global optimum point. The method is most applicable when the number of available training samples is small. The performance of the proposed localized method is successfully demonstrated on a large variety of data sets. We demonstrate that the number of features selected by the lLFS method saturates at the number of available discriminative features. In addition, we have shown that the Vapnik-Chervonenkis dimension of the localized classifier is finite. Both these factors suggest that the lLFS method is insensitive to the overfitting issue, relative to other methods. [ABSTRACT FROM AUTHOR]

As the main performance self-evaluation index of the Kalman filter, the estimation error covariance (EEC) has been used to design the allocation cost function of task and resources for sensor tracking networks. For nonlinear systems, the sensor allocation method based on the EEC needs to adjust the allocation plans after obtaining the filtering results. Meanwhile, recent investigations have indicated that the self-evaluation function EEC of the Kalman filtering is universally inapplicable in practical applications, for which the estimation models are generally mismatched due to difficulty in accurately training parameters and approximation of nonlinear systems. Thereby, the sensors cannot be properly allocated by using the EEC as a preliminary criterion. Alternatively, observable degree (OD) is a naturally quantitative measure on observability and can be utilized to effectively measure the estimation performance. In this paper, the OD analysis with scale transform invariance for nonlinear systems is studied by using the unscented Kalman filter, the pseudostate transition matrix, and the pseudo observation matrix on the basis of the results of linear systems. Afterward, the OD of nonlinear fusion systems, the sensor utilization efficiency, the priority of tasks, and the sensor performance and sensitivity are jointly considered to formulate the optimization problem for sensor allocation. The genetic algorithm with intelligent learning function is employed to solve the optimization problem. Moreover, extensive simulation demonstrates the feasibility of the proposed approach. [ABSTRACT FROM AUTHOR]

Collaboration enables weak species to survive in an environment where different species compete for limited resources. Cooperative coevolution (CC) is a nature-inspired optimization method that divides a problem into subcomponents and evolves them while genetically isolating them. Problem decomposition is an important aspect in using CC for neuroevolution. CC employs different problem decomposition methods to decompose the neural network training problem into subcomponents. Different problem decomposition methods have features that are helpful at different stages in the evolutionary process. Adaptation, collaboration, and competition are needed for CC, as multiple subpopulations are used to represent the problem. It is important to add collaboration and competition in CC. This paper presents a competitive CC method for training recurrent neural networks for chaotic time-series prediction. Two different instances of the competitive method are proposed that employs different problem decomposition methods to enforce island-based competition. The results show improvement in the performance of the proposed methods in most cases when compared with standalone CC and other methods from the literature. [ABSTRACT FROM PUBLISHER]

This paper deals with the design of an H {\infty } tracking controller for nonlinear continuous-time systems with completely unknown dynamics. A general bounded L2 -gain tracking problem with a discounted performance function is introduced for the H {\infty } tracking. A tracking Hamilton–Jacobi–Isaac (HJI) equation is then developed that gives a Nash equilibrium solution to the associated min–max optimization problem. A rigorous analysis of bounded L2 -gain and stability of the control solution obtained by solving the tracking HJI equation is provided. An upper-bound is found for the discount factor to assure local asymptotic stability of the tracking error dynamics. An off-policy reinforcement learning algorithm is used to learn the solution to the tracking HJI equation online without requiring any knowledge of the system dynamics. Convergence of the proposed algorithm to the solution to the tracking HJI equation is shown. Simulation examples are provided to verify the effectiveness of the proposed method. [ABSTRACT FROM PUBLISHER]

Wafer defect detection using an intelligent system is an approach of quality improvement in semiconductor manufacturing that aims to enhance its process stability, increase production capacity, and improve yields. Occasionally, only few records that indicate defective units are available and they are classified as a minority group in a large database. Such a situation leads to an imbalanced data set problem, wherein it engenders a great challenge to deal with by applying machine-learning techniques for obtaining effective solution. In addition, the database may comprise overlapping samples of different classes. This paper introduces two models of evolutionary fuzzy ARTMAP (FAM) neural networks to deal with the imbalanced data set problems in a semiconductor manufacturing operations. In particular, both the FAM models and hybrid genetic algorithms are integrated in the proposed evolutionary artificial neural networks (EANNs) to classify an imbalanced data set. In addition, one of the proposed EANNs incorporates a facility to learn overlapping samples of different classes from the imbalanced data environment. The classification results of the proposed evolutionary FAM neural networks are presented, compared, and analyzed using several classification metrics. The outcomes positively indicate the effectiveness of the proposed networks in handling classification problems with imbalanced data sets. [ABSTRACT FROM PUBLISHER]

The multivalued neuron with periodic activation function (MVN-P) was proposed by Aizenberg for solving classification problems. The boundaries between two distinct categories are crisply specified in MVN-P, which may result in slow convergence or being unable to converge at all in the learning process. In this paper, we propose a revised model of MVN-P based on the idea of unsharp boundaries. In this revised model, a fuzzy buffer is provided around a boundary between two distinct categories, allowing incorrect assignments with membership degree less than a threshold to be tolerated in the training phase. Genetic algorithms are applied to derive optimal values for the parameters involved in this model, alleviating the burden of setting them manually by the user. Besides, MVN-P has difficulties solving the classification problems having a large number of categories. A tree structure is developed to overcome these difficulties. Simulation results demonstrate the effectiveness of our proposed ideas. [ABSTRACT FROM AUTHOR]

An accurate and noninvasive stress assessment from human physiology is a strenuous task. In this paper, a pattern recognition system to learn complex correlates between heart rate variability (HRV) features and salivary stress biomarkers is proposed. Using the Trier social stress test, heart rate and salivary measurements were obtained from volunteers under varying levels of stress induction. Measurements of salivary alpha-amylase and cortisol were used as objective measures of stress, and were correlated with the HRV features using fuzzy ARTMAP (FAM). In improving the predictive ability of the ARTMAPs, techniques, such as genetic algorithms for parameter optimization and voting ensembles, were employed. The ensemble of FAMs can be used for predicting stress responses of salivary alpha-amylase or cortisol using heart rate measurements as the input. Using alpha-amylase as the stress indicator, the ensemble was able to classify stress from heart rate features with 75% accuracy, and 80% accuracy when cortisol was used. [ABSTRACT FROM AUTHOR]

The performance of convolutional neural networks (CNNs) highly relies on their architectures. In order to design a CNN with promising performance, extensive expertise in both CNNs and the investigated problem domain is required, which is not necessarily available to every interested user. To address this problem, we propose to automatically evolve CNN architectures by using a genetic algorithm (GA) based on ResNet and DenseNet blocks. The proposed algorithm is completely automatic in designing CNN architectures. In particular, neither preprocessing before it starts nor postprocessing in terms of CNNs is needed. Furthermore, the proposed algorithm does not require users with domain knowledge on CNNs, the investigated problem, or even GAs. The proposed algorithm is evaluated on the CIFAR10 and CIFAR100 benchmark data sets against 18 state-of-the-art peer competitors. Experimental results show that the proposed algorithm outperforms the state-of-the-art CNNs hand-crafted and the CNNs designed by automatic peer competitors in terms of the classification performance and achieves a competitive classification accuracy against semiautomatic peer competitors. In addition, the proposed algorithm consumes much less computational resource than most peer competitors in finding the best CNN architectures. [ABSTRACT FROM AUTHOR]

Modeling of distributed parameter systems is difficult because of their nonlinearity and infinite-dimensional characteristics. Based on principal component analysis (PCA), a hybrid modeling strategy that consists of a decoupled linear autoregressive exogenous (ARX) model and a nonlinear radial basis function (RBF) neural network model are proposed. The spatial-temporal output is first divided into a few dominant spatial basis functions and finite-dimensional temporal series by PCA. Then, a decoupled ARX model is designed to model the linear dynamics of the dominant modes of the time series. The nonlinear residual part is subsequently parameterized by RBFs, where genetic algorithm is utilized to optimize their hidden layer structure and the parameters. Finally, the nonlinear spatial-temporal dynamic system is obtained after the time/space reconstruction. Simulation results of a catalytic rod and a heat conduction equation demonstrate the effectiveness of the proposed strategy compared to several other methods. [ABSTRACT FROM AUTHOR]

In this brief, an enhanced genetic back-propagation neural network with link switches (EGA-BPNN-LS) is proposed to address a data-driven modeling problem for gasification processes inside United Gas Improvement (UGI) gasifiers. The online-measured temperature of crude gas produced during the gasification processes plays a dominant role in the syngas industry; however, it is difficult to model temperature dynamics via first principles due to the practical complexity of the gasification process, especially as reflected by severe changes in the gas temperature resulting from infrequent manipulations of the gasifier in practice. The proposed data-driven modeling approach, EGA-BPNN-LS, incorporates an NN-LS, an EGA, and the Levenberg-Marquardt (LM) algorithm. The approach cannot only learn the relationships between the control input and the system output from historical data using an optimized network structure through a combination of EGA and NN-LS but also makes use of the networks gradient information via the LM algorithm. EGA-BPNN-LS is applied to a set of data collected from the field to model the UGI gasification processes, and the effectiveness of EGA-BPNN-LS is verified. [ABSTRACT FROM PUBLISHER]

We consider the task of performing prediction with neural networks (NNs) on the basis of uncertain input data expressed in the form of intervals. We aim at quantifying the uncertainty in the prediction arising from both the input data and the prediction model. A multilayer perceptron NN is trained to map interval-valued input data onto interval outputs, representing the prediction intervals (PIs) of the real target values. The NN training is performed by nondominated sorting genetic algorithm-II, so that the PIs are optimized both in terms of accuracy (coverage probability) and dimension (width). Demonstration of the proposed method is given in two case studies: 1) a synthetic case study, in which the data have been generated with a 5-min time frequency from an autoregressive moving average model with either Gaussian or Chi-squared innovation distribution and 2) a real case study, in which experimental data consist of wind speed measurements with a time step of 1 h. Comparisons are given with a crisp (single-valued) approach. The results show that the crisp approach is less reliable than the interval-valued input approach in terms of capturing the variability in input. [ABSTRACT FROM PUBLISHER]