Artificial Neural Network Learning Algorithms Research Articles

Seafloor sediment classification from single beam echo sounder data using LVQ network

Seafloor sediment classification based on echo characteristics obtained from single-beam echosounder is very useful in remote and instant sediment classification. Results of different classification techniques using such data provide robust results when the acoustic beam has a normal incidence with the seabottom. This may not always be true and show poor classification, with the data acquired during rough sea periods corresponding to both oblique and normal incidence of the acoustic pulse, due to roll and pitch motion of the ship. In the present study, an attempt is made to exploit the artificial neural network (ANN) techniques for better classification with such data. Learning Vector Quantisation (LVQ) is a supervised learning algorithm of ANN that is found to be an effective tool and show good performance. The input data to the network include the roughness index (E1) and hardness index (E2) derived from echo characteristics. The network utilizes the competitive learning, a distance function in the first layer and a linear function in the second layer. The network was tried with a different size of hidden neurons and training data size to see the influence on classification. It is found that with ten neurons in the first layer and four neurons in the second layer good performance in classification for the data was achieved.

Dimensionality Reduction of near Infrared Spectral Data Using Global and Local Implementations of Principal Component Analysis for Neural Network Calibrations

The artificial neural network (ANN) learning algorithm is a strong alternative to traditional linear calibration methods used in near infrared (NIR) spectroscopy. The generalisation capacity of ANN can be effectively employed only when the ratio of available training samples to a number of neuron interconnection weights and biases (unknown regression parameters) is large. Traditionally, this ratio is increased by compressing X data to fewer variables using principal component analysis (PCA). An alternative method, local PCA compression, has been described in the literature. This approach overcomes the global linearity of PCA by performing dimensionality reduction in two steps: division of the data space into clusters and local compression of each cluster using PCA. This study compared global and local implementations of PCA compression in NIR calibration problems solved with ANN regression. Three data sets were used for development of control (based on PCA) and experimental (based on local PCA) ANN calibrations. The results demonstrated that local PCA could outperform traditional global PCA compression. However, the best dimensionality reduction method was case-dependent. Performance of local PCA-based calibrations degraded rapidly as compression rate increased, while global PCA allowed higher compression at minimal cost of prediction accuracy.

Bayesian neural network for rainfall‐runoff modeling

In this paper, a Bayesian learning approach is introduced to train a multilayer feed‐forward network for daily river flow and reservoir inflow simulation in a cold region river basin in Canada. In Bayesian approach, uncertainty about the relationship between inputs and outputs is initially taken care of by an assumed prior distribution of parameters (weights and biases). This prior distribution is updated to posterior distribution using a likelihood function following Bayes' theorem while data are observed. This posterior distribution is called the objective function of a network in the Bayesian learning approach. The objective function is maximized using a suitable optimization technique. Once the network is trained, the predictive distribution of the network outputs is obtained by integrating over the posterior distribution of weights. In this study, Gaussian prior distribution and a Gaussian noise model are used in defining posterior distribution. The network has been optimized using a scaled conjugate gradient technique. Posterior distribution of weights is approximated to Gaussian during prediction. Prediction performance of the Bayesian neural network (BNN) is compared with the results obtained from a standard artificial neural network (ANN) model and a widely used conceptual rainfall‐runoff model, namely, HBV‐96. The BNN model outperformed the conceptual model and slightly outperformed the standard ANN model in simulating mean, peak, and low river flows and reservoir inflows. The significant contribution of the Bayesian method over the conventional ANN approach, among others, is the uncertainty estimation of the outputs in the form of confidence intervals which are particularly needed in practical water resources applications. Prediction confidence limits (or intervals) indicate the extent to which one can rely on predictions for decision making. It is shown that the BNN can provide reliable streamflow and reservoir inflow forecasts without a loss in model prediction accuracy as compared to standard ANN and conceptual model HBV. Another significant advantage of BNN approach is that the overfitting and underfitting problems are automatically taken care of by the Bayesian learning algorithm, which conversely remain serious problems with conventional ANN learning algorithm.

Implementasi Jaringan Syaraf Tiruan Untuk Mengidentifikasi Pola Desain Produk Berdasarkan Preferensi Pelanggan Menggunakan Kansei Engineering System

Kansei Engineering System (KES) is of technologies in ergonomics that support human-centered design. In Japanese, Kansei means customers’ taste to a certain product. The research aims to apply KES Type II (a more recent development of KES)with artificial neural network (back propagation model) and traingdx learning algorithm to identify product design that fits best with customers’ taste/preferences.The model developed is able to generate a design for steering wheel with various atributes; namely shape of button, number of spokes, material of spokes, texture of grip, style of grip, and color. These atributes are results from 16 pairs of Kansei words which is reduced into five main factors; namely aestethics, emotional evaluation, color, texture, and product reliability. Keywords: Kansei Engineering System, artificial neural networks, back-propagation, customers’ preferences.

Two-Phase Pattern Search-based Learning Method for Multi-layer Neural Network

A new multi-layer artificial neural network learning algorithm based on pattern search method is proposed. The learning model has two phases-a pattern search phase, and a local minimum-escaping phase. In the pattern search phase, our method performs local search iteratively and minimize the error measure function along with the set of descent directions of the error measure directly and finds the nearest minima efficiently. When the network gets stuck in local minima, the local minimum-escaping phase attempts to fill up the valley by modifying temperature parameters in ascent direction of the error measure. Thus, the two phases are repeated until the network gets out of local minima. The learning model is designed to provide a very simple and effective means of searching the minima of objective function directly without any knowledge of its derivatives. We test this algorithm on benchmark problems, such as exclusive-or (XOR), parity, Arabic numerals recognition, function approximation problems and a real world classification task. For all problems, the systems are shown be trained efficiently by our method. As a simple direct search method, it can be applied in hardware implementations easily.

Acoustic and auditory representations of the voicing contrast

Key aspects of the categorical perception of initial stop consonants by human and animal listeners are replicated by a ‘‘synthetic listener’’ in the form of a computational auditory model. This model is a composite of a physiologically faithful ‘‘front end’’ [M. J. Pont and R. I. Damper, J. Acoust. Soc. Am. 89, 1213–1228 (1991)] and an artificial neural network (ANN) ‘‘back end.’’ ANNs are trained on neurograms (i.e., the front end’s auditory-nerve representation) for the 0- and 80-ms end points of three voice-onset time continua (bilabial, alveolar, and velar), and tested on the intermediate (10, 20, ..., 70 ms) patterns. Replication of the behavioral data—in terms of warping of similarity space and shift of phoneme boundary with place of articulation—is robust in that it is maintained across a variety of ANN architectures and learning algorithms. Unlike real listeners, the synthetic listener can be analyzed easily and manipulated purposefully to reveal the acoustic and auditory features underlying the voicing contrast. A contribution analysis of the weights and activations of the ANNs highlights the auditory features. Replacement of the Pont–Damper front end by a simple Fourier-based spectral analysis abolishes the boundary-shift effect, although similarity warping is maintained. This suggests that restructuring of the acoustic features by the peripheral auditory system is crucial to phonetic classification.

Nuclear Power Plant Status Diagnostics Using an Artificial Neural Network

In this work, nuclear power plant operating status recognition is investigated using a self-optimizing stochastic learning algorithm artificial neural network (ANN) with dynamic node architecture learning. The objective is to train the ANN to classify selected nuclear power plant accident conditions and assess the potential for future success in this area. The network is trained on normal operating conditions as well as on potentially unsafe conditions based on nuclear power plant training simulator-generated accident scenarios. These scenarios include hot- and cold-leg loss of coolant, control rod ejection, total loss of off-site power, main steamline break, main feedwater line break, and steam generator tube leak accidents as well as the normal operating condition. Findings show that ANNs can be used to diagnose and classify nuclear power plant conditions with good results. Continued research work is indicated.

Design of low-cost, real-time simulation systems for large neural networks

Systems with large amounts of computing power and storage are required to simulate very large neural networks capable of tackling complex control problems and real-time emulation of the human sensory, language, and reasoning systems. General-purpose parallel computers do not have communications, processor, and memory architectures optimized for neural computation and so cannot perform such simulations at reasonable cost. This paper analyzes several software and hardware strategies to make feasible the simulation of large neural networks in real-time and presents a particular multicomputer design able to implement these strategies. An important design goal is that the system must not sacrifice computational flexibility for speed as new information about the workings of the brain and new artificial neural network architectures and learning algorithms are continually emerging.

Neural networks—Their applications and perspectives in intelligent machining

In intelligent manufacturing systems, unprecedented und unforeseen situations are expected to be solved, within certain limits, even on the basis of incomplete and inprecise information One has tried to achieve this goal since years. Gradually, it seems to be realizable through partial solutions, integrated in today's flexible manufacturing systems. These complexes are fairly complicated material and data processing systems, in which different sensors and actuators are thoroughly distributed. The most important requirements for the intelligent techniques to be applied in these systems are the abilities for integration of multiple sensor information, for real-time functioning, for effective knowledge representation and for learning or adaptivity. Artificial neural networks seem to be able to fulfil some of these requirements and to contribute to the implementation of partial solutions, which can lead to the realization of truly intelligent manufacturing systems. The paper gives a summary of known neural networks applications and perspectives in intelligent manufacturing. Special emphasis is given on intelligent machining, namely on the following fields: multisensor fusion and integration, learning of process models, adaptive control, monitoring, diagnostics and quality control. For the sake of perspicuity a short survey of different artificial neural network structures and learning algorithms is also given, together with frequent applications of neural network techniques in fields different from intelligent manufacturing.

Multiprocessor systems for connectionist diagnosis of technical processes

The paper summarizes the requirements for a new family of monitoring and diagnostic systems, characterized among others by networking capability, high computing speed and adaptive learning ability. The diagnostic process is regarded as a pattern recognition procedure. Artificial neural networks (ANNs) or connectionist models and their pattern recognition abilities are illustrated. The back propagation technique as the most frequently used learning algorithm for multi-layered ANNs is outlined together with some acceleration methods. Neural network forms of traditional pattern recognition approaches are described, which usually mean a direct determination of the networks' parameters. The paper gives the first results of investigations comparing the learning and recognition performance of frequently used traditional pattern recognition techniques, back propagation networks and a network based on the condensed nearest-neighbour concept, developed at the University of Paderborn. Two “neuro monitoring and diagnostic systems” based on parallel processor structures and incorporating neural network techniques for learning and classification, being developed at the Institute for Electrical Measurement, University of Paderborn, are presented and compared in the paper.