A computational study on the performance of artificial neural networks under changing structural design and data distribution

This paper provides the results of our computational studies on artificial neural networks (ANNs) under various structural design and data distributions. A two-group classification problem is investigated. Simulated data with varying kurtosis and variance are used to examine how the ANN performs with respect to certain structural design (size and addition of input and weight noise) characteristics. The results of our study indicate that additive noise, size, and data distribution characteristics play an important role in learning, reliability and predictive ability of ANNs.     

Using artificial neural networks to solve the orienteering problem

In the orienteering problem, we are given a transportation network in which a start point and an end point are specified. Other points have associated scores. Given a fixed amount of time, the goal is to determine a path from start to end through a subset of locations in order to maximize the total path score. This problem has received a considerable amount of attention in the last ten years. The traveling salesman problem is a variant of the orienteering problem. This paper applies a modified, continuous Hopfield neural network to attack this NP-hard optimization problem. In it, we design an effective energy function and learning algorithm. Unlike some applications of neural networks to optimization problems, this approach is shown to perform quite well.     

Structural evaluation of materials by artificial neural networks

It is shown that due to the complexity of interaction of the excitation field with a material in determining its physical characteristics, as well as sophisticated correlation relationships between the physical characteristics and structure of a real material, in many cases, relization of the structural evaluation of materials on the basis of idealized mathematical models of the underlying physical processes is of limited use. As an alternative, it is proposed to use an artificial neural network for the extraction of quantitative information of interest from measurements of the physical characteristics. A general overview of artificial neural networks is given. A methodology of the use of a multilayer perceptron for determining various structural parameters from the dielectric spectra is described. As an example, determination of the moisture content and density of sawdust from the dielectric modulusis considered by the neural-network technique. The noise performance of the neural network is analyzed by applying an additive and multiplicative noise to the input data.Institute of Polymer Mechanics, University of Latvia, Riga, LV-1006 Latvia. Published in Mekhanika Kompozitnykh Materialov, Vol. 35, No. 1, pp. 109–124, January–February, 1999.     

Path relinking and GRG for artificial neural networks

Artificial neural networks (ANN) have been widely used for both classification and prediction. This paper is focused on the prediction problem in which an unknown function is approximated. ANNs can be viewed as models of real systems, built by tuning parameters known as weights. In training the net, the problem is to find the weights that optimize its performance (i.e., to minimize the error over the training set). Although the most popular method for training these networks is back propagation, other optimization methods such as tabu search or scatter search have been successfully applied to solve this problem. In this paper we propose a path relinking implementation to solve the neural network training problem. Our method uses GRG, a gradient-based local NLP solver, as an improvement phase, while previous approaches used simpler local optimizers. The experimentation shows that the proposed procedure can compete with the best-known algorithms in terms of solution quality, consuming a reasonable computational effort.     

Design of fuzzy radial basis function-based polynomial neural networks

In this study, we introduce a new design methodology of fuzzy radial basis function-based polynomial neural networks. In many cases, these models do not come with capabilities to deal with granular information. With this regard, fuzzy sets offer several interesting and useful opportunities. This study presents the development of fuzzy radial basis function-based neural networks augmented with virtual input variables. The performance of the proposed category of models is quantified through a series of experiments, in which we use two machine learning data sets and two publicly available software development effort data.     

Modelling public transport trips by radial basis function neural networks

Artificial neural networks (ANNs) are one of the recently explored advanced technologies, which show promise in the area of transportation engineering. The presented study used two different ANN algorithms, feed forward back-propagation (FFBP) and radial basis function (RBF), for the purpose of daily trip flow forecasting. The ANN predictions were quite close to the observations as reflected in the selected performance criteria. The selected stochastic model performance was quite poor compared with ANN results. It was seen that the RBF neural network did not provide negative forecasts in contrast to FFBP applications. Besides, the local minima problem faced by some FFBP algorithms was not encountered in RBF networks.     

Modeling of thermotransport phenomenon in metal alloys using artificial neural networks

Thermodiffusion in molten metals, also known as thermotransport, a phenomenon in which constituent elements of an alloy separate under the influence of non-uniform temperature field, is of significance in several applications. However, due to the complex inter-particle interactions, there is no theoretical formulation that can model this phenomenon with adequate accuracy. Keeping in mind the severe deficiencies of the present day thermotransport models and an urgent need of a reliable method in several engineering applications ranging from crystal growth to integrated circuit design to nuclear reactor designs, an engineering approach has been taken in which neurocomputing principles have been employed to develop artificial neural network models to study and quantify the thermotransport phenomenon in binary metal alloys. Unlike any other thermotransport model for molten metals, the neural network approach has been validated for several types of binary alloys, viz., concentrated, dilute, isotopic and non-isotopic metals. Additionally, to establish the soundness of the model and to highlight its potential as a unified computational analysis tool, it ability to capture several thermotransport trends has been shown. Comparison with other models from the literature has also been made indicating a superior performance of this technique with respect to several other well established thermotransport models.     

Thermal behaviour prediction utilizing artificial neural networks for an open office

This study investigates a neural network-based non-linear autoregressive model with external inputs (NNARX), a non-linear autoregressive moving average model with external inputs (NNARMAX), and a non-linear output error model (NNOE) to predict the thermal behaviour of an open-plan office in a modern commercial building. External and internal climate data recorded over one summer, autumn and winter season were used to build and validate the models. The paper illustrates the potential of using these models to predict room temperature and relative humidity for different time scales ahead (30 min or 2 h ahead). The prediction performance is evaluated using the criteria of goodness of fit, coefficient of determination, mean absolute error and mean squared error between predicted model output and real measurements. To obtain an optimal network structure (avoiding overfitting) after training, a pruning algorithm called optimal brain surgeon (OBS) was used to remove unnecessary input signals, weights and hidden neurons. The results demonstrate that all models provide reasonably good predictions but the NNARX and NNARMAX models outperform the NNOE model. These models can all potentially be used for improving the performance of thermal environment control systems.     

Hybrid artificial neural networks for efficient valuation of real options and financial derivatives

A hybrid valuation methodology is proposed and tested for improving the efficiency of contingent claims pricing by combining Artificial Neural Networks (ANN) and conventional parametric option pricing techniques. With one application on financial derivatives and one on real options the methods superiority is demonstrated. The resulting efficiency is instrumental for real time applications.MSC code: 90-08 Acknowledgements: Both authors are thankful for partial financial support to the HERMES European Center of Excellence on Computational Finance and Economics of the University of Cyprus and a University of Cyprus grant for research in ANNs and Derivatives, and to the anonymous referees for their helpful comments and discussions.     

Exponential stability preservation in discrete-time analogues of artificial neural networks with distributed delays

This paper demonstrates that there is a discrete-time analogue which does not require any restriction on the size of the time-step in order to preserve the exponential stability of an artificial neural network with distributed delays. The analysis exploits an appropriate Lyapunov sequence and a discrete-time system of Halanay inequalities, and also either a Young inequality or a geometric-arithmetic mean inequality, to derive several sufficient conditions on the network parameters for the exponential stability of the analogue. The sufficiency conditions are independent of the time-step, and they correspond to those that establish the exponential stability of the continuous-time network.     

The rate of approximation of Gaussian radial basis neural networks in continuous function space

There have been many studies on the dense theorem of approximation by radial basis feedforword neural networks, and some approximation problems by Gaussian radial basis feedforward neural networks(GRBFNs)in some special function space have also been investigated. This paper considers the approximation by the GRBFNs in continuous function space. It is proved that the rate of approximation by GRNFNs with n~d neurons to any continuous function f defined on a compact subset K(R~d)can be controlled by ω(f, n~(-1/2)), where ω(f, t)is the modulus of continuity of the function f .     

Post-training on RBF neural networks

Radial basis function neural networks are the most widely used networks due to their rapid training, generality, and simplicity. The nature of these networks necessitates some types of errors which can never be removed by traditional training algorithms. This paper is an attempt to introduce the natural error sources of neural networks such as bias error, iteration-restricted error, and Gibbs error. Moreover, a new method is introduced, called post-training, to reduce these errors as far as desired.     

The three-factor model and artificial neural networks: predicting stock price movement in China

Since the establishment of the Shanghai Stock Exchange (SHSE) in 1990 and the Shenzhen Stock Exchange (SZSE) in 1991, China’s stock markets have expanded rapidly. Although this rapid growth has attracted considerable academic interest, few studies have examined the ability of conventional financial models to predict the share price movements of Chinese stock. This gap in the literature is significant, given the volatility of the Chinese stock markets and the added risk that arises from the Chinese legal and regulatory environment. In this paper we address this research gap by examining the predictive ability of several well-established forecasting models, including dynamic versions of a single-factor CAPM-based model and Fama and French’s three-factor model. In addition, we compare the forecasting ability of each of these models with that of an artificial neural network (ANN) model that contains the same predictor variables but relaxes the assumption of model linearity. Surprisingly, we find no statistical differences in the forecasting accuracy of the CAPM and three-factor model, a result that may reflect the emerging nature of the Chinese stock markets. We also find that each ANN model outperforms the corresponding linear model, indicating that neural networks may be a useful tool for stock price prediction in emerging markets.     

A mended hybrid learning algorithm for radial basis function neural networks to improve generalization capability

A two-step learning scheme for radial basis function neural networks, which combines the genetic algorithm (GA) with the hybrid learning algorithm (HLA), is proposed in this paper. It is compared with the methods of the GA, the recursive orthogonal least square algorithm (ROLSA) and another two-step learning scheme for RBF neural networks, which combines the K-means clustering with the HLA (K-means + HLA). Our proposed method has the best generalization performance.     

Application of artificial neural networks for the prediction of roll force and roll torque in hot strip rolling process

This paper introduces an artificial neural network (ANN) application to a hot strip mill to improve the model’s prediction ability for rolling force and rolling torque, as a function of various process parameters. To obtain a data basis for training and validation of the neural network, numerous three dimensional finite element simulations were carried out for different sets of process variables. Experimental data were compared with the finite element predictions to verify the model accuracy. The input variables are selected to be rolling speed, percentage of thickness reduction, initial temperature of the strip and friction coefficient in the contact area. A comprehensive analysis of the prediction errors of roll force and roll torque made by the ANN is presented. Model responses analysis is also conducted to enhance the understanding of the behavior of the NN model. The resulted ANN model is feasible for on-line control and rolling schedule optimization, and can be easily extended to cover different aluminum grades and strip sizes in a straight-forward way by generating the corresponding training data from a FE model.