A two-layer recurrent neural network for nonsmooth convex optimization problems.

Science.gov (United States)

Qin, Sitian; Xue, Xiaoping

2015-06-01

In this paper, a two-layer recurrent neural network is proposed to solve the nonsmooth convex optimization problem subject to convex inequality and linear equality constraints. Compared with existing neural network models, the proposed neural network has a low model complexity and avoids penalty parameters. It is proved that from any initial point, the state of the proposed neural network reaches the equality feasible region in finite time and stays there thereafter. Moreover, the state is unique if the initial point lies in the equality feasible region. The equilibrium point set of the proposed neural network is proved to be equivalent to the Karush-Kuhn-Tucker optimality set of the original optimization problem. It is further proved that the equilibrium point of the proposed neural network is stable in the sense of Lyapunov. Moreover, from any initial point, the state is proved to be convergent to an equilibrium point of the proposed neural network. Finally, as applications, the proposed neural network is used to solve nonlinear convex programming with linear constraints and L1 -norm minimization problems.

Numerical Analysis of Modeling Based on Improved Elman Neural Network

Directory of Open Access Journals (Sweden)

Shao Jie

2014-01-01

Full Text Available A modeling based on the improved Elman neural network (IENN is proposed to analyze the nonlinear circuits with the memory effect. The hidden layer neurons are activated by a group of Chebyshev orthogonal basis functions instead of sigmoid functions in this model. The error curves of the sum of squared error (SSE varying with the number of hidden neurons and the iteration step are studied to determine the number of the hidden layer neurons. Simulation results of the half-bridge class-D power amplifier (CDPA with two-tone signal and broadband signals as input have shown that the proposed behavioral modeling can reconstruct the system of CDPAs accurately and depict the memory effect of CDPAs well. Compared with Volterra-Laguerre (VL model, Chebyshev neural network (CNN model, and basic Elman neural network (BENN model, the proposed model has better performance.

Liver Tumor Segmentation from MR Images Using 3D Fast Marching Algorithm and Single Hidden Layer Feedforward Neural Network

Directory of Open Access Journals (Sweden)

Trong-Ngoc Le

2016-01-01

Full Text Available Objective. Our objective is to develop a computerized scheme for liver tumor segmentation in MR images. Materials and Methods. Our proposed scheme consists of four main stages. Firstly, the region of interest (ROI image which contains the liver tumor region in the T1-weighted MR image series was extracted by using seed points. The noise in this ROI image was reduced and the boundaries were enhanced. A 3D fast marching algorithm was applied to generate the initial labeled regions which are considered as teacher regions. A single hidden layer feedforward neural network (SLFN, which was trained by a noniterative algorithm, was employed to classify the unlabeled voxels. Finally, the postprocessing stage was applied to extract and refine the liver tumor boundaries. The liver tumors determined by our scheme were compared with those manually traced by a radiologist, used as the “ground truth.” Results. The study was evaluated on two datasets of 25 tumors from 16 patients. The proposed scheme obtained the mean volumetric overlap error of 27.43% and the mean percentage volume error of 15.73%. The mean of the average surface distance, the root mean square surface distance, and the maximal surface distance were 0.58 mm, 1.20 mm, and 6.29 mm, respectively.

Cardiac Arrhythmia Classification by Multi-Layer Perceptron and Convolution Neural Networks

Directory of Open Access Journals (Sweden)

Shalin Savalia

2018-05-01

Full Text Available The electrocardiogram (ECG plays an imperative role in the medical field, as it records heart signal over time and is used to discover numerous cardiovascular diseases. If a documented ECG signal has a certain irregularity in its predefined features, this is called arrhythmia, the types of which include tachycardia, bradycardia, supraventricular arrhythmias, and ventricular, etc. This has encouraged us to do research that consists of distinguishing between several arrhythmias by using deep neural network algorithms such as multi-layer perceptron (MLP and convolution neural network (CNN. The TensorFlow library that was established by Google for deep learning and machine learning is used in python to acquire the algorithms proposed here. The ECG databases accessible at PhysioBank.com and kaggle.com were used for training, testing, and validation of the MLP and CNN algorithms. The proposed algorithm consists of four hidden layers with weights, biases in MLP, and four-layer convolution neural networks which map ECG samples to the different classes of arrhythmia. The accuracy of the algorithm surpasses the performance of the current algorithms that have been developed by other cardiologists in both sensitivity and precision.

Cardiac Arrhythmia Classification by Multi-Layer Perceptron and Convolution Neural Networks.

Science.gov (United States)

Savalia, Shalin; Emamian, Vahid

2018-05-04

The electrocardiogram (ECG) plays an imperative role in the medical field, as it records heart signal over time and is used to discover numerous cardiovascular diseases. If a documented ECG signal has a certain irregularity in its predefined features, this is called arrhythmia, the types of which include tachycardia, bradycardia, supraventricular arrhythmias, and ventricular, etc. This has encouraged us to do research that consists of distinguishing between several arrhythmias by using deep neural network algorithms such as multi-layer perceptron (MLP) and convolution neural network (CNN). The TensorFlow library that was established by Google for deep learning and machine learning is used in python to acquire the algorithms proposed here. The ECG databases accessible at PhysioBank.com and kaggle.com were used for training, testing, and validation of the MLP and CNN algorithms. The proposed algorithm consists of four hidden layers with weights, biases in MLP, and four-layer convolution neural networks which map ECG samples to the different classes of arrhythmia. The accuracy of the algorithm surpasses the performance of the current algorithms that have been developed by other cardiologists in both sensitivity and precision.

Discriminative training of self-structuring hidden control neural models

DEFF Research Database (Denmark)

Sørensen, Helge Bjarup Dissing; Hartmann, Uwe; Hunnerup, Preben

1995-01-01

This paper presents a new training algorithm for self-structuring hidden control neural (SHC) models. The SHC models were trained non-discriminatively for speech recognition applications. Better recognition performance can generally be achieved, if discriminative training is applied instead. Thus...... we developed a discriminative training algorithm for SHC models, where each SHC model for a specific speech pattern is trained with utterances of the pattern to be recognized and with other utterances. The discriminative training of SHC neural models has been tested on the TIDIGITS database...

Implementation of multi-layer feed forward neural network on PIC16F877 microcontroller

International Nuclear Information System (INIS)

Nur Aira Abd Rahman

2005-01-01

Artificial Neural Network (ANN) is an electronic model based on the neural structure of the brain. Similar to human brain, ANN consists of interconnected simple processing units or neurons that process input to generate output signals. ANN operation is divided into 2 categories; training mode and service mode. This project aims to implement ANN on PIC micro-controller that enable on-chip or stand alone training and service mode. The input can varies from sensors or switches, while the output can be used to control valves, motors, light source and a lot more. As partial development of the project, this paper reports the current status and results of the implemented ANN. The hardware fraction of this project incorporates Microchip PIC16F877A microcontrollers along with uM-FPU math co-processor. uM-FPU is a 32-bit floating point co-processor utilized to execute complex calculation requires by the sigmoid activation function for neuron. ANN algorithm is converted to software program written in assembly language. The implemented ANN structure is three layer with one hidden layer, and five neurons with two hidden neurons. To prove the operability and functionality, the network is trained to solve three common logic gate operations; AND, OR, and XOR. This paper concludes that the ANN had been successfully implemented on PIC16F877a and uM-FPU math co-processor hardware that works accordingly on both training and service mode. (Author)

Optimized hardware framework of MLP with random hidden layers for classification applications

Science.gov (United States)

Zyarah, Abdullah M.; Ramesh, Abhishek; Merkel, Cory; Kudithipudi, Dhireesha

2016-05-01

Multilayer Perceptron Networks with random hidden layers are very efficient at automatic feature extraction and offer significant performance improvements in the training process. They essentially employ large collection of fixed, random features, and are expedient for form-factor constrained embedded platforms. In this work, a reconfigurable and scalable architecture is proposed for the MLPs with random hidden layers with a customized building block based on CORDIC algorithm. The proposed architecture also exploits fixed point operations for area efficiency. The design is validated for classification on two different datasets. An accuracy of ~ 90% for MNIST dataset and 75% for gender classification on LFW dataset was observed. The hardware has 299 speed-up over the corresponding software realization.

LEARNING ALGORITHM EFFECT ON MULTILAYER FEED FORWARD ARTIFICIAL NEURAL NETWORK PERFORMANCE IN IMAGE CODING

Directory of Open Access Journals (Sweden)

OMER MAHMOUD

2007-08-01

Full Text Available One of the essential factors that affect the performance of Artificial Neural Networks is the learning algorithm. The performance of Multilayer Feed Forward Artificial Neural Network performance in image compression using different learning algorithms is examined in this paper. Based on Gradient Descent, Conjugate Gradient, Quasi-Newton techniques three different error back propagation algorithms have been developed for use in training two types of neural networks, a single hidden layer network and three hidden layers network. The essence of this study is to investigate the most efficient and effective training methods for use in image compression and its subsequent applications. The obtained results show that the Quasi-Newton based algorithm has better performance as compared to the other two algorithms.

Autonomous Navigation Apparatus With Neural Network for a Mobile Vehicle

Science.gov (United States)

Quraishi, Naveed (Inventor)

1996-01-01

An autonomous navigation system for a mobile vehicle arranged to move within an environment includes a plurality of sensors arranged on the vehicle and at least one neural network including an input layer coupled to the sensors, a hidden layer coupled to the input layer, and an output layer coupled to the hidden layer. The neural network produces output signals representing respective positions of the vehicle, such as the X coordinate, the Y coordinate, and the angular orientation of the vehicle. A plurality of patch locations within the environment are used to train the neural networks to produce the correct outputs in response to the distances sensed.

Aero Engine Component Fault Diagnosis Using Multi-Hidden-Layer Extreme Learning Machine with Optimized Structure

Directory of Open Access Journals (Sweden)

Shan Pang

2016-01-01

Full Text Available A new aero gas turbine engine gas path component fault diagnosis method based on multi-hidden-layer extreme learning machine with optimized structure (OM-ELM was proposed. OM-ELM employs quantum-behaved particle swarm optimization to automatically obtain the optimal network structure according to both the root mean square error on training data set and the norm of output weights. The proposed method is applied to handwritten recognition data set and a gas turbine engine diagnostic application and is compared with basic ELM, multi-hidden-layer ELM, and two state-of-the-art deep learning algorithms: deep belief network and the stacked denoising autoencoder. Results show that, with optimized network structure, OM-ELM obtains better test accuracy in both applications and is more robust to sensor noise. Meanwhile it controls the model complexity and needs far less hidden nodes than multi-hidden-layer ELM, thus saving computer memory and making it more efficient to implement. All these advantages make our method an effective and reliable tool for engine component fault diagnosis tool.

Tuning Recurrent Neural Networks for Recognizing Handwritten Arabic Words

KAUST Repository

Qaralleh, Esam

2013-10-01

Artificial neural networks have the abilities to learn by example and are capable of solving problems that are hard to solve using ordinary rule-based programming. They have many design parameters that affect their performance such as the number and sizes of the hidden layers. Large sizes are slow and small sizes are generally not accurate. Tuning the neural network size is a hard task because the design space is often large and training is often a long process. We use design of experiments techniques to tune the recurrent neural network used in an Arabic handwriting recognition system. We show that best results are achieved with three hidden layers and two subsampling layers. To tune the sizes of these five layers, we use fractional factorial experiment design to limit the number of experiments to a feasible number. Moreover, we replicate the experiment configuration multiple times to overcome the randomness in the training process. The accuracy and time measurements are analyzed and modeled. The two models are then used to locate network sizes that are on the Pareto optimal frontier. The approach described in this paper reduces the label error from 26.2% to 19.8%.

Feature to prototype transition in neural networks

Science.gov (United States)

Krotov, Dmitry; Hopfield, John

Models of associative memory with higher order (higher than quadratic) interactions, and their relationship to neural networks used in deep learning are discussed. Associative memory is conventionally described by recurrent neural networks with dynamical convergence to stable points. Deep learning typically uses feedforward neural nets without dynamics. However, a simple duality relates these two different views when applied to problems of pattern classification. From the perspective of associative memory such models deserve attention because they make it possible to store a much larger number of memories, compared to the quadratic case. In the dual description, these models correspond to feedforward neural networks with one hidden layer and unusual activation functions transmitting the activities of the visible neurons to the hidden layer. These activation functions are rectified polynomials of a higher degree rather than the rectified linear functions used in deep learning. The network learns representations of the data in terms of features for rectified linear functions, but as the power in the activation function is increased there is a gradual shift to a prototype-based representation, the two extreme regimes of pattern recognition known in cognitive psychology. Simons Center for Systems Biology.

Neural Network Models for Free Radical Polymerization of Methyl Methacrylate

International Nuclear Information System (INIS)

Curteanu, S.; Leon, F.; Galea, D.

2003-01-01

In this paper, a neural network modeling of the batch bulk methyl methacrylate polymerization is performed. To obtain conversion, number and weight average molecular weights, three neural networks were built. Each was a multilayer perception with one or two hidden layers. The choice of network topology, i.e. the number of hidden layers and the number of neurons in these layers, was based on achieving a compromise between precision and complexity. Thus, it was intended to have an error as small as possible at the end of back-propagation training phases, while using a network with reduced complexity. The performances of the networks were evaluated by comparing network predictions with training data, validation data (which were not uses for training), and with the results of a mechanistic model. The accurate predictions of neural networks for monomer conversion, number average molecular weight and weight average molecular weight proves that this modeling methodology gives a good representation and generalization of the batch bulk methyl methacrylate polymerization. (author)

Chaos Synchronization Using Adaptive Dynamic Neural Network Controller with Variable Learning Rates

Directory of Open Access Journals (Sweden)

Chih-Hong Kao

2011-01-01

Full Text Available This paper addresses the synchronization of chaotic gyros with unknown parameters and external disturbance via an adaptive dynamic neural network control (ADNNC system. The proposed ADNNC system is composed of a neural controller and a smooth compensator. The neural controller uses a dynamic RBF (DRBF network to online approximate an ideal controller. The DRBF network can create new hidden neurons online if the input data falls outside the hidden layer and prune the insignificant hidden neurons online if the hidden neuron is inappropriate. The smooth compensator is designed to compensate for the approximation error between the neural controller and the ideal controller. Moreover, the variable learning rates of the parameter adaptation laws are derived based on a discrete-type Lyapunov function to speed up the convergence rate of the tracking error. Finally, the simulation results which verified the chaotic behavior of two nonlinear identical chaotic gyros can be synchronized using the proposed ADNNC scheme.

Assessing the effect of quantitative and qualitative predictors on gastric cancer individuals survival using hierarchical artificial neural network models.

Science.gov (United States)

Amiri, Zohreh; Mohammad, Kazem; Mahmoudi, Mahmood; Parsaeian, Mahbubeh; Zeraati, Hojjat

2013-01-01

There are numerous unanswered questions in the application of artificial neural network models for analysis of survival data. In most studies, independent variables have been studied as qualitative dichotomous variables, and results of using discrete and continuous quantitative, ordinal, or multinomial categorical predictive variables in these models are not well understood in comparison to conventional models. This study was designed and conducted to examine the application of these models in order to determine the survival of gastric cancer patients, in comparison to the Cox proportional hazards model. We studied the postoperative survival of 330 gastric cancer patients who suffered surgery at a surgical unit of the Iran Cancer Institute over a five-year period. Covariates of age, gender, history of substance abuse, cancer site, type of pathology, presence of metastasis, stage, and number of complementary treatments were entered in the models, and survival probabilities were calculated at 6, 12, 18, 24, 36, 48, and 60 months using the Cox proportional hazards and neural network models. We estimated coefficients of the Cox model and the weights in the neural network (with 3, 5, and 7 nodes in the hidden layer) in the training group, and used them to derive predictions in the study group. Predictions with these two methods were compared with those of the Kaplan-Meier product limit estimator as the gold standard. Comparisons were performed with the Friedman and Kruskal-Wallis tests. Survival probabilities at different times were determined using the Cox proportional hazards and a neural network with three nodes in the hidden layer; the ratios of standard errors with these two methods to the Kaplan-Meier method were 1.1593 and 1.0071, respectively, revealed a significant difference between Cox and Kaplan-Meier (P neural network, and the neural network and the standard (Kaplan-Meier), as well as better accuracy for the neural network (with 3 nodes in the hidden layer

Random neural Q-learning for obstacle avoidance of a mobile robot in unknown environments

Directory of Open Access Journals (Sweden)

Jing Yang

2016-07-01

Full Text Available The article presents a random neural Q-learning strategy for the obstacle avoidance problem of an autonomous mobile robot in unknown environments. In the proposed strategy, two independent modules, namely, avoidance without considering the target and goal-seeking without considering obstacles, are first trained using the proposed random neural Q-learning algorithm to obtain their best control policies. Then, the two trained modules are combined based on a switching function to realize the obstacle avoidance in unknown environments. For the proposed random neural Q-learning algorithm, a single-hidden layer feedforward network is used to approximate the Q-function to estimate the Q-value. The parameters of the single-hidden layer feedforward network are modified using the recently proposed neural algorithm named the online sequential version of extreme learning machine, where the parameters of the hidden nodes are assigned randomly and the sample data can come one by one. However, different from the original online sequential version of extreme learning machine algorithm, the initial output weights are estimated subjected to quadratic inequality constraint to improve the convergence speed. Finally, the simulation results demonstrate that the proposed random neural Q-learning strategy can successfully solve the obstacle avoidance problem. Also, the higher learning efficiency and better generalization ability are achieved by the proposed random neural Q-learning algorithm compared with the Q-learning based on the back-propagation method.

Failure detection studies by layered neural network

International Nuclear Information System (INIS)

Ciftcioglu, O.; Seker, S.; Turkcan, E.

1991-06-01

Failure detection studies by layered neural network (NN) are described. The particular application area is an operating nuclear power plant and the failure detection is of concern as result of system surveillance in real-time. The NN system is considered to be consisting of 3 layers, one of which being hidden, and the NN parameters are determined adaptively by the backpropagation (BP) method, the process being the training phase. Studies are performed using the power spectra of the pressure signal of the primary system of an operating nuclear power plant of PWR type. The studies revealed that, by means of NN approach, failure detection can effectively be carried out using the redundant information as well as this is the case in this work; namely, from measurement of the primary pressure signals one can estimate the primary system coolant temperature and hence the deviation from the operational temperature state, the operational status identified in the training phase being referred to as normal. (author). 13 refs.; 4 figs.; 2 tabs

Design of a universal two-layered neural network derived from the PLI theory

Science.gov (United States)

Hu, Chia-Lun J.

2004-05-01

The if-and-only-if (IFF) condition that a set of M analog-to-digital vector-mapping relations can be learned by a one-layered-feed-forward neural network (OLNN) is that all the input analog vectors dichotomized by the i-th output bit must be positively, linearly independent, or PLI. If they are not PLI, then the OLNN just cannot learn no matter what learning rules is employed because the solution of the connection matrix does not exist mathematically. However, in this case, one can still design a parallel-cascaded, two-layered, perceptron (PCTLP) to acheive this general mapping goal. The design principle of this "universal" neural network is derived from the major mathematical properties of the PLI theory - changing the output bits of the dependent relations existing among the dichotomized input vectors to make the PLD relations PLI. Then with a vector concatenation technique, the required mapping can still be learned by this PCTLP system with very high efficiency. This paper will report in detail the mathematical derivation of the general design principle and the design procedures of the PCTLP neural network system. It then will be verified in general by a practical numerical example.

IMPLEMENTASI BACKPROPAGATION NEURAL NETWORK DALAM PRAKIRAAN CUACA DI DAERAH BALI SELATAN

Directory of Open Access Journals (Sweden)

I MADE DWI UDAYANA PUTRA

2016-11-01

Full Text Available Weather information has an important role in human life in various fields, such as agriculture, marine, and aviation. The accurate weather forecasts are needed in order to improve the performance of various fields. In this study, use artificial neural network method with backpropagation learning algorithm to create a model of weather forecasting in the area of ??South Bali. The aim of this study is to determine the effect of the number of neurons in the hidden layer and to determine the level of accuracy of the method of artificial neural network with backpropagation learning algorithm in weather forecast models. Weather forecast models in this study use input of the factors that influence the weather, namely air temperature, dew point, wind speed, visibility, and barometric pressure.The results of testing the network with a different number of neurons in the hidden layer of artificial neural network method with backpropagation learning algorithms show that the increase in the number of neurons in the hidden layer is not directly proportional to the value of the accuracy of the weather forecasts, the increase in the number of neurons in the hidden layer does not necessarily increase or decrease value accuracy of weather forecasts we obtain the best accuracy rate of 51.6129% on a network model with three neurons in the hidden layer.

TeraHertz imaging of hidden paint layers on canvas

NARCIS (Netherlands)

Adam, A.J.L.; Planken, P.C.M.; Meloni, S.; Dik, J.

2009-01-01

We show terahertz reflection images of hidden paint layers in a painting on canvas and compare the results with X-ray Radiography and Infrared Reflectography. Our terahertz measurements show strong reflections from both the canvas/paint interface and from the raw umber/lead white interface,

Temperature, Humidity and Energy Consumption Forecasting in the Poultry Hall Using Artificial Neural Networknetwork

Directory of Open Access Journals (Sweden)

N Gholamrezaei

2017-10-01

trainlm algorithm (Levenberg-Marquardt was used. To simulate temperature, humidity and energy consumption, networks were trained with two and three layers, respectively. Network with two layers with10 neurons in the hidden layer and one neuron in the output layer with (R² equal to 0.96 and (MSE equal to 0.00912, was given the best result for predicting temperature. For humidity electronic sensors, results showed that network with three layers with the 10 neurons in the first hidden layer, 20 neurons in the second hidden layer and one neuron in the output layer with (R² equal to 0.8 and (MSE equal to 0.00783 was the best for predicting humidity. Finally, network with two layers with 10 neurons in the first hidden layer, 10 neurons in the second hidden layer and one neuron in the output layer was selected as the optimal structure for predicting energy consumption. For this topology, (R² and MSE were determined to 0.98 and 0.00114, respectively. Linear and multivariate regression for the parameters affecting temperature, humidity and energy consumption of electronic sensors was determined by the STATGR software. Correlation coefficients indicated that parameters such as length, height and width of the electronic control sensors placed in the poultry hall justified 82% of the temperature changes, 61% of the humidity changes and 92% of the energy consumption changes. Therefore, comparing with correlation coefficients obtained from the neural network models, the highest correlation coefficient was related to energy parameter and the lowest correlation was linked to humidity parameter. Conclusions The results of the study indicated the high performance for predicting temperature, humidity and energy consumption. The networks hadthree inputs including length, width and height of electronic sensor positions and an output for temperature, humidity and energy consumption. For training networks the multiple layer perceptron (MLP with error back propagation learning algorithm (BP

Improved algorithms for circuit fault diagnosis based on wavelet packet and neural network

International Nuclear Information System (INIS)

Zhang, W-Q; Xu, C

2008-01-01

In this paper, two improved BP neural network algorithms of fault diagnosis for analog circuit are presented through using optimal wavelet packet transform(OWPT) or incomplete wavelet packet transform(IWPT) as preprocessor. The purpose of preprocessing is to reduce the nodes in input layer and hidden layer of BP neural network, so that the neural network gains faster training and convergence speed. At first, we apply OWPT or IWPT to the response signal of circuit under test(CUT), and then calculate the normalization energy of each frequency band. The normalization energy is used to train the BP neural network to diagnose faulty components in the analog circuit. These two algorithms need small network size, while have faster learning and convergence speed. Finally, simulation results illustrate the two algorithms are effective for fault diagnosis

A neural network model for credit risk evaluation.

Science.gov (United States)

Khashman, Adnan

2009-08-01

Credit scoring is one of the key analytical techniques in credit risk evaluation which has been an active research area in financial risk management. This paper presents a credit risk evaluation system that uses a neural network model based on the back propagation learning algorithm. We train and implement the neural network to decide whether to approve or reject a credit application, using seven learning schemes and real world credit applications from the Australian credit approval datasets. A comparison of the system performance under the different learning schemes is provided, furthermore, we compare the performance of two neural networks; with one and two hidden layers following the ideal learning scheme. Experimental results suggest that neural networks can be effectively used in automatic processing of credit applications.

Prediction of Cascading Collapse Occurrence due to the Effect of Hidden Failure of a Protection System using Artificial Neural Network

Directory of Open Access Journals (Sweden)

Nor Hazwani Idris

2017-06-01

Full Text Available Transmission line act as a medium of transportation for electrical energy from a power station to the consumer. There are many factors that could cause the cascading collapse such as instability of voltage and frequency, the change of environment and weather, the software and operator error and also the failure in protection system. Protection system plays an important function in maintaining the stability and reliability of the power grid. Hidden failures in relay protection systems are the primary factors for triggering the cascading collapse. This paper presents an Artificial Neural Network (ANN model for prediction of cascading collapse occurrence due to the effect of hidden failure of protection system. The ANN model has been developed through the normalized training and testing data process with optimum number of hidden layer, the momentum rate and the learning rate. The ANN model employs probability of hidden failure, random number of line limit power flow and exposed line as its input while trip index of cascading collapse occurrence as its output. IEEE 14 bus system is used in this study to illustrate the proposed approach. The performance of the results is analysed in terms of its Mean Square Error (MSE and Correlation Coefficient (R. The results show the ANN model produce reliable prediction of cascading collapse occurrence.

Fundamental study on the interpretation technique for 3-D MT data using neural networks. 2; Neural network wo mochiita sanjigen MT ho data kaishaku gijutsu ni kansuru kisoteki kenkyu. 2

Energy Technology Data Exchange (ETDEWEB)

Fukuoka, K; Kobayashi, T [OYO Corp., Tokyo (Japan); Mogi, T [Kyushu University, Fukuoka (Japan). Faculty of Engineering; Spichak, V

1997-10-22

Behavior of neural networks relative to noise and the constitution of an optimum network are studied for the construction of a 3-D MT data interpretation system using neural networks. In the study, the relationship is examined between the noise level of educational data and the noise level of the neural network to be constructed. After examination it is found that the neural network is effective in interpreting data whose noise level is the same as that of educational data; it cannot correctly interpret data that it has not met in the educational stage even if such data is free of noise; that the optimum number of neurons in a hidden layer is approximately 40 in a network architecture using the current system; and that the neuron gain function enhances recognition capability when a logistic function is used in the hidden layer and a linear function is used in the output layer. 2 refs., 7 figs., 2 tabs.

Direct and inverse neural networks modelling applied to study the influence of the gas diffusion layer properties on PBI-based PEM fuel cells

Energy Technology Data Exchange (ETDEWEB)

Lobato, Justo; Canizares, Pablo; Rodrigo, Manuel A.; Linares, Jose J. [Chemical Engineering Department, University of Castilla-La Mancha, Campus Universitario s/n, 13004 Ciudad Real (Spain); Piuleac, Ciprian-George; Curteanu, Silvia [Faculty of Chemical Engineering and Environmental Protection, Department of Chemical Engineering, ' ' Gh. Asachi' ' Technical University Iasi Bd. D. Mangeron, No. 71A, 700050 IASI (Romania)

2010-08-15

This article shows the application of a very useful mathematical tool, artificial neural networks, to predict the fuel cells results (the value of the tortuosity and the cell voltage, at a given current density, and therefore, the power) on the basis of several properties that define a Gas Diffusion Layer: Teflon content, air permeability, porosity, mean pore size, hydrophobia level. Four neural networks types (multilayer perceptron, generalized feedforward network, modular neural network, and Jordan-Elman neural network) have been applied, with a good fitting between the predicted and the experimental values in the polarization curves. A simple feedforward neural network with one hidden layer proved to be an accurate model with good generalization capability (error about 1% in the validation phase). A procedure based on inverse neural network modelling was able to determine, with small errors, the initial conditions leading to imposed values for characteristics of the fuel cell. In addition, the use of this tool has been proved to be very attractive in order to predict the cell performance, and more interestingly, the influence of the properties of the gas diffusion layer on the cell performance, allowing possible enhancements of this material by changing some of its properties. (author)

Folk music style modelling by recurrent neural networks with long short term memory units

OpenAIRE

Sturm, Bob; Santos, João Felipe; Korshunova, Iryna

2015-01-01

We demonstrate two generative models created by training a recurrent neural network (RNN) with three hidden layers of long short-term memory (LSTM) units. This extends past work in numerous directions, including training deeper models with nearly 24,000 high-level transcriptions of folk tunes. We discuss our on-going work.

Synchronization and Inter-Layer Interactions of Noise-Driven Neural Networks.

Science.gov (United States)

Yuniati, Anis; Mai, Te-Lun; Chen, Chi-Ming

2017-01-01

In this study, we used the Hodgkin-Huxley (HH) model of neurons to investigate the phase diagram of a developing single-layer neural network and that of a network consisting of two weakly coupled neural layers. These networks are noise driven and learn through the spike-timing-dependent plasticity (STDP) or the inverse STDP rules. We described how these networks transited from a non-synchronous background activity state (BAS) to a synchronous firing state (SFS) by varying the network connectivity and the learning efficacy. In particular, we studied the interaction between a SFS layer and a BAS layer, and investigated how synchronous firing dynamics was induced in the BAS layer. We further investigated the effect of the inter-layer interaction on a BAS to SFS repair mechanism by considering three types of neuron positioning (random, grid, and lognormal distributions) and two types of inter-layer connections (random and preferential connections). Among these scenarios, we concluded that the repair mechanism has the largest effect for a network with the lognormal neuron positioning and the preferential inter-layer connections.

Comparing the Selected Transfer Functions and Local Optimization Methods for Neural Network Flood Runoff Forecast

Directory of Open Access Journals (Sweden)

Petr Maca

2014-01-01

Full Text Available The presented paper aims to analyze the influence of the selection of transfer function and training algorithms on neural network flood runoff forecast. Nine of the most significant flood events, caused by the extreme rainfall, were selected from 10 years of measurement on small headwater catchment in the Czech Republic, and flood runoff forecast was investigated using the extensive set of multilayer perceptrons with one hidden layer of neurons. The analyzed artificial neural network models with 11 different activation functions in hidden layer were trained using 7 local optimization algorithms. The results show that the Levenberg-Marquardt algorithm was superior compared to the remaining tested local optimization methods. When comparing the 11 nonlinear transfer functions, used in hidden layer neurons, the RootSig function was superior compared to the rest of analyzed activation functions.

Using Elman recurrent neural networks with conjugate gradient algorithm in determining the anesthetic the amount of anesthetic medicine to be applied.

Science.gov (United States)

Güntürkün, Rüştü

2010-08-01

In this study, Elman recurrent neural networks have been defined by using conjugate gradient algorithm in order to determine the depth of anesthesia in the continuation stage of the anesthesia and to estimate the amount of medicine to be applied at that moment. The feed forward neural networks are also used for comparison. The conjugate gradient algorithm is compared with back propagation (BP) for training of the neural Networks. The applied artificial neural network is composed of three layers, namely the input layer, the hidden layer and the output layer. The nonlinear activation function sigmoid (sigmoid function) has been used in the hidden layer and the output layer. EEG data has been recorded with Nihon Kohden 9200 brand 22-channel EEG device. The international 8-channel bipolar 10-20 montage system (8 TB-b system) has been used in assembling the recording electrodes. EEG data have been recorded by being sampled once in every 2 milliseconds. The artificial neural network has been designed so as to have 60 neurons in the input layer, 30 neurons in the hidden layer and 1 neuron in the output layer. The values of the power spectral density (PSD) of 10-second EEG segments which correspond to the 1-50 Hz frequency range; the ratio of the total power of PSD values of the EEG segment at that moment in the same range to the total of PSD values of EEG segment taken prior to the anesthesia.

Resolution of Singularities Introduced by Hierarchical Structure in Deep Neural Networks.

Science.gov (United States)

Nitta, Tohru

2017-10-01

We present a theoretical analysis of singular points of artificial deep neural networks, resulting in providing deep neural network models having no critical points introduced by a hierarchical structure. It is considered that such deep neural network models have good nature for gradient-based optimization. First, we show that there exist a large number of critical points introduced by a hierarchical structure in deep neural networks as straight lines, depending on the number of hidden layers and the number of hidden neurons. Second, we derive a sufficient condition for deep neural networks having no critical points introduced by a hierarchical structure, which can be applied to general deep neural networks. It is also shown that the existence of critical points introduced by a hierarchical structure is determined by the rank and the regularity of weight matrices for a specific class of deep neural networks. Finally, two kinds of implementation methods of the sufficient conditions to have no critical points are provided. One is a learning algorithm that can avoid critical points introduced by the hierarchical structure during learning (called avoidant learning algorithm). The other is a neural network that does not have some critical points introduced by the hierarchical structure as an inherent property (called avoidant neural network).

Single-Iteration Learning Algorithm for Feed-Forward Neural Networks

Energy Technology Data Exchange (ETDEWEB)

Barhen, J.; Cogswell, R.; Protopopescu, V.

1999-07-31

A new methodology for neural learning is presented, whereby only a single iteration is required to train a feed-forward network with near-optimal results. To this aim, a virtual input layer is added to the multi-layer architecture. The virtual input layer is connected to the nominal input layer by a specird nonlinear transfer function, and to the fwst hidden layer by regular (linear) synapses. A sequence of alternating direction singular vrdue decompositions is then used to determine precisely the inter-layer synaptic weights. This algorithm exploits the known separability of the linear (inter-layer propagation) and nonlinear (neuron activation) aspects of information &ansfer within a neural network.

Novel maximum-margin training algorithms for supervised neural networks.

Science.gov (United States)

Ludwig, Oswaldo; Nunes, Urbano

2010-06-01

This paper proposes three novel training methods, two of them based on the backpropagation approach and a third one based on information theory for multilayer perceptron (MLP) binary classifiers. Both backpropagation methods are based on the maximal-margin (MM) principle. The first one, based on the gradient descent with adaptive learning rate algorithm (GDX) and named maximum-margin GDX (MMGDX), directly increases the margin of the MLP output-layer hyperplane. The proposed method jointly optimizes both MLP layers in a single process, backpropagating the gradient of an MM-based objective function, through the output and hidden layers, in order to create a hidden-layer space that enables a higher margin for the output-layer hyperplane, avoiding the testing of many arbitrary kernels, as occurs in case of support vector machine (SVM) training. The proposed MM-based objective function aims to stretch out the margin to its limit. An objective function based on Lp-norm is also proposed in order to take into account the idea of support vectors, however, overcoming the complexity involved in solving a constrained optimization problem, usually in SVM training. In fact, all the training methods proposed in this paper have time and space complexities O(N) while usual SVM training methods have time complexity O(N (3)) and space complexity O(N (2)) , where N is the training-data-set size. The second approach, named minimization of interclass interference (MICI), has an objective function inspired on the Fisher discriminant analysis. Such algorithm aims to create an MLP hidden output where the patterns have a desirable statistical distribution. In both training methods, the maximum area under ROC curve (AUC) is applied as stop criterion. The third approach offers a robust training framework able to take the best of each proposed training method. The main idea is to compose a neural model by using neurons extracted from three other neural networks, each one previously trained by

Selection of Hidden Layer Neurons and Best Training Method for FFNN in Application of Long Term Load Forecasting

Science.gov (United States)

Singh, Navneet K.; Singh, Asheesh K.; Tripathy, Manoj

2012-05-01

For power industries electricity load forecast plays an important role for real-time control, security, optimal unit commitment, economic scheduling, maintenance, energy management, and plant structure planning etc. A new technique for long term load forecasting (LTLF) using optimized feed forward artificial neural network (FFNN) architecture is presented in this paper, which selects optimal number of neurons in the hidden layer as well as the best training method for the case study. The prediction performance of proposed technique is evaluated using mean absolute percentage error (MAPE) of Thailand private electricity consumption and forecasted data. The results obtained are compared with the results of classical auto-regressive (AR) and moving average (MA) methods. It is, in general, observed that the proposed method is prediction wise more accurate.

Sequential neural models with stochastic layers

DEFF Research Database (Denmark)

Fraccaro, Marco; Sønderby, Søren Kaae; Paquet, Ulrich

2016-01-01

How can we efficiently propagate uncertainty in a latent state representation with recurrent neural networks? This paper introduces stochastic recurrent neural networks which glue a deterministic recurrent neural network and a state space model together to form a stochastic and sequential neural...... generative model. The clear separation of deterministic and stochastic layers allows a structured variational inference network to track the factorization of the model's posterior distribution. By retaining both the nonlinear recursive structure of a recurrent neural network and averaging over...

Neural network for recognizing signal-shape of nuclear detector output

International Nuclear Information System (INIS)

Mardiyanto M Panitra

2006-01-01

The use of artificial intelligent technique in the engineering field has been familiar especially in the field of pattern recognition. By using this technique, either simple routine works or complicated routine works can be done by the help of a digital camera and a personal computer. One of the complicated works that can not be solved easily is how to separate two kinds of nuclear radiation types which are mixed in the same field. The separation of the two kinds of radiation become is very important for the radiation dosimetry purposes. For doing this we have carried out a preliminary research in applying a neural network technique for recognizing C and T letters with right, left, up, and down positions. We arranged a three-layer neural network i.e. input layer (9 neurons with/without bias neuron), hidden layer (11 neurons), and output layer (1 neuron). From this preliminary study the use of a bias neuron gave faster learning process compared with the one without the bias neuron. The neural network could work successfully in determining the letter S and T without any mistake. (author)

Estimation of effective connectivity using multi-layer perceptron artificial neural network.

Science.gov (United States)

Talebi, Nasibeh; Nasrabadi, Ali Motie; Mohammad-Rezazadeh, Iman

2018-02-01

Studies on interactions between brain regions estimate effective connectivity, (usually) based on the causality inferences made on the basis of temporal precedence. In this study, the causal relationship is modeled by a multi-layer perceptron feed-forward artificial neural network, because of the ANN's ability to generate appropriate input-output mapping and to learn from training examples without the need of detailed knowledge of the underlying system. At any time instant, the past samples of data are placed in the network input, and the subsequent values are predicted at its output. To estimate the strength of interactions, the measure of " Causality coefficient " is defined based on the network structure, the connecting weights and the parameters of hidden layer activation function. Simulation analysis demonstrates that the method, called "CREANN" (Causal Relationship Estimation by Artificial Neural Network), can estimate time-invariant and time-varying effective connectivity in terms of MVAR coefficients. The method shows robustness with respect to noise level of data. Furthermore, the estimations are not significantly influenced by the model order (considered time-lag), and the different initial conditions (initial random weights and parameters of the network). CREANN is also applied to EEG data collected during a memory recognition task. The results implicate that it can show changes in the information flow between brain regions, involving in the episodic memory retrieval process. These convincing results emphasize that CREANN can be used as an appropriate method to estimate the causal relationship among brain signals.

Character Recognition Using Genetically Trained Neural Networks

Energy Technology Data Exchange (ETDEWEB)

Diniz, C.; Stantz, K.M.; Trahan, M.W.; Wagner, J.S.

1998-10-01

Computationally intelligent recognition of characters and symbols addresses a wide range of applications including foreign language translation and chemical formula identification. The combination of intelligent learning and optimization algorithms with layered neural structures offers powerful techniques for character recognition. These techniques were originally developed by Sandia National Laboratories for pattern and spectral analysis; however, their ability to optimize vast amounts of data make them ideal for character recognition. An adaptation of the Neural Network Designer soflsvare allows the user to create a neural network (NN_) trained by a genetic algorithm (GA) that correctly identifies multiple distinct characters. The initial successfid recognition of standard capital letters can be expanded to include chemical and mathematical symbols and alphabets of foreign languages, especially Arabic and Chinese. The FIN model constructed for this project uses a three layer feed-forward architecture. To facilitate the input of characters and symbols, a graphic user interface (GUI) has been developed to convert the traditional representation of each character or symbol to a bitmap. The 8 x 8 bitmap representations used for these tests are mapped onto the input nodes of the feed-forward neural network (FFNN) in a one-to-one correspondence. The input nodes feed forward into a hidden layer, and the hidden layer feeds into five output nodes correlated to possible character outcomes. During the training period the GA optimizes the weights of the NN until it can successfully recognize distinct characters. Systematic deviations from the base design test the network's range of applicability. Increasing capacity, the number of letters to be recognized, requires a nonlinear increase in the number of hidden layer neurodes. Optimal character recognition performance necessitates a minimum threshold for the number of cases when genetically training the net. And, the

Anti-pairing in learning of a neural network with redundant hidden units

International Nuclear Information System (INIS)

Kwon, Chulan; Kim, Hyong Kyun

2005-01-01

We study the statistical mechanics of learning from examples between the two-layered committee machines with different numbers of hidden units using the replica theory. The number M of hidden units of the student network is larger than the number M T of those of the target network called the teacher. We choose the networks to have binary synaptic weights, ±1, which makes it possible to compare the calculation with the Monte Carlo simulation. We propose an effective teacher as a virtual target network which has the same M hidden units as the student and gives identical outputs with those of the original teacher. This is a way of making a conjecture for a ground state of a thermodynamic system, given by the weights of the effective teacher in our study. We suppose that the weights on M T hidden units of the effective teacher are the same as those of the original teacher while those on M - M T redundant hidden units are composed of anti-pairs, {1, - 1}, with probability 1 - p in the limit p → 0. For p = 0 exact, there are no terms related to the effective teacher in the calculation, for the contributions of anti-pairs to outputs are exactly cancelled. In the limit p → 0, however, we find that the learnt weights of the student are actually equivalent to those of the suggested effective teacher, which is not possible from the calculation for p = 0. p plays the role of a symmetry breaking parameter for anti-pairing ordering, which is analogous to the magnetic field for the Ising model. A first-order phase transition is found to be signalled by breaking of symmetry in permuting hidden units. Above a critical number of examples, the student is shown to learn perfectly the effective teacher. Anti-pairing can be measured by a set of order parameters; zero in the permutation-symmetric phase and nonzero in the permutation symmetry breaking phase. Results from the Monte Carlo simulation are shown to be in good agreement with those from the replica calculation

Universal approximation in p-mean by neural networks

NARCIS (Netherlands)

Burton, R.M; Dehling, H.G

A feedforward neural net with d input neurons and with a single hidden layer of n neurons is given by [GRAPHICS] where a(j), theta(j), w(ji) is an element of R. In this paper we study the approximation of arbitrary functions f: R-d --> R by a neural net in an L-p(mu) norm for some finite measure mu

Evolving RBF neural networks for adaptive soft-sensor design.

Science.gov (United States)

Alexandridis, Alex

2013-12-01

This work presents an adaptive framework for building soft-sensors based on radial basis function (RBF) neural network models. The adaptive fuzzy means algorithm is utilized in order to evolve an RBF network, which approximates the unknown system based on input-output data from it. The methodology gradually builds the RBF network model, based on two separate levels of adaptation: On the first level, the structure of the hidden layer is modified by adding or deleting RBF centers, while on the second level, the synaptic weights are adjusted with the recursive least squares with exponential forgetting algorithm. The proposed approach is tested on two different systems, namely a simulated nonlinear DC Motor and a real industrial reactor. The results show that the produced soft-sensors can be successfully applied to model the two nonlinear systems. A comparison with two different adaptive modeling techniques, namely a dynamic evolving neural-fuzzy inference system (DENFIS) and neural networks trained with online backpropagation, highlights the advantages of the proposed methodology.

Deep neural mapping support vector machines.

Science.gov (United States)

Li, Yujian; Zhang, Ting

2017-09-01

The choice of kernel has an important effect on the performance of a support vector machine (SVM). The effect could be reduced by NEUROSVM, an architecture using multilayer perceptron for feature extraction and SVM for classification. In binary classification, a general linear kernel NEUROSVM can be theoretically simplified as an input layer, many hidden layers, and an SVM output layer. As a feature extractor, the sub-network composed of the input and hidden layers is first trained together with a virtual ordinary output layer by backpropagation, then with the output of its last hidden layer taken as input of the SVM classifier for further training separately. By taking the sub-network as a kernel mapping from the original input space into a feature space, we present a novel model, called deep neural mapping support vector machine (DNMSVM), from the viewpoint of deep learning. This model is also a new and general kernel learning method, where the kernel mapping is indeed an explicit function expressed as a sub-network, different from an implicit function induced by a kernel function traditionally. Moreover, we exploit a two-stage procedure of contrastive divergence learning and gradient descent for DNMSVM to jointly training an adaptive kernel mapping instead of a kernel function, without requirement of kernel tricks. As a whole of the sub-network and the SVM classifier, the joint training of DNMSVM is done by using gradient descent to optimize the objective function with the sub-network layer-wise pre-trained via contrastive divergence learning of restricted Boltzmann machines. Compared to the separate training of NEUROSVM, the joint training is a new algorithm for DNMSVM to have advantages over NEUROSVM. Experimental results show that DNMSVM can outperform NEUROSVM and RBFSVM (i.e., SVM with the kernel of radial basis function), demonstrating its effectiveness. Copyright © 2017 Elsevier Ltd. All rights reserved.

Germ layers, the neural crest and emergent organization in development and evolution.

Science.gov (United States)

Hall, Brian K

2018-04-10

Discovered in chick embryos by Wilhelm His in 1868 and named the neural crest by Arthur Milnes Marshall in 1879, the neural crest cells that arise from the neural folds have since been shown to differentiate into almost two dozen vertebrate cell types and to have played major roles in the evolution of such vertebrate features as bone, jaws, teeth, visceral (pharyngeal) arches, and sense organs. I discuss the discovery that ectodermal neural crest gave rise to mesenchyme and the controversy generated by that finding; the germ layer theory maintained that only mesoderm could give rise to mesenchyme. A second topic of discussion is germ layers (including the neural crest) as emergent levels of organization in animal development and evolution that facilitated major developmental and evolutionary change. The third topic is gene networks, gene co-option, and the evolution of gene-signaling pathways as key to developmental and evolutionary transitions associated with the origin and evolution of the neural crest and neural crest cells. © 2018 Wiley Periodicals, Inc.

Artificial neural network intelligent method for prediction

Science.gov (United States)

Trifonov, Roumen; Yoshinov, Radoslav; Pavlova, Galya; Tsochev, Georgi

2017-09-01

Accounting and financial classification and prediction problems are high challenge and researchers use different methods to solve them. Methods and instruments for short time prediction of financial operations using artificial neural network are considered. The methods, used for prediction of financial data as well as the developed forecasting system with neural network are described in the paper. The architecture of a neural network used four different technical indicators, which are based on the raw data and the current day of the week is presented. The network developed is used for forecasting movement of stock prices one day ahead and consists of an input layer, one hidden layer and an output layer. The training method is algorithm with back propagation of the error. The main advantage of the developed system is self-determination of the optimal topology of neural network, due to which it becomes flexible and more precise The proposed system with neural network is universal and can be applied to various financial instruments using only basic technical indicators as input data.

Tuning Recurrent Neural Networks for Recognizing Handwritten Arabic Words

KAUST Repository

Qaralleh, Esam; Abandah, Gheith; Jamour, Fuad Tarek

2013-01-01

and sizes of the hidden layers. Large sizes are slow and small sizes are generally not accurate. Tuning the neural network size is a hard task because the design space is often large and training is often a long process. We use design of experiments

A stochastic learning algorithm for layered neural networks

International Nuclear Information System (INIS)

Bartlett, E.B.; Uhrig, R.E.

1992-01-01

The random optimization method typically uses a Gaussian probability density function (PDF) to generate a random search vector. In this paper the random search technique is applied to the neural network training problem and is modified to dynamically seek out the optimal probability density function (OPDF) from which to select the search vector. The dynamic OPDF search process, combined with an auto-adaptive stratified sampling technique and a dynamic node architecture (DNA) learning scheme, completes the modifications of the basic method. The DNA technique determines the appropriate number of hidden nodes needed for a given training problem. By using DNA, researchers do not have to set the neural network architectures before training is initiated. The approach is applied to networks of generalized, fully interconnected, continuous perceptions. Computer simulation results are given

Accurate prediction of the dew points of acidic combustion gases by using an artificial neural network model

International Nuclear Information System (INIS)

ZareNezhad, Bahman; Aminian, Ali

2011-01-01

This paper presents a new approach based on using an artificial neural network (ANN) model for predicting the acid dew points of the combustion gases in process and power plants. The most important acidic combustion gases namely, SO 3 , SO 2 , NO 2 , HCl and HBr are considered in this investigation. Proposed Network is trained using the Levenberg-Marquardt back propagation algorithm and the hyperbolic tangent sigmoid activation function is applied to calculate the output values of the neurons of the hidden layer. According to the network's training, validation and testing results, a three layer neural network with nine neurons in the hidden layer is selected as the best architecture for accurate prediction of the acidic combustion gases dew points over wide ranges of acid and moisture concentrations. The proposed neural network model can have significant application in predicting the condensation temperatures of different acid gases to mitigate the corrosion problems in stacks, pollution control devices and energy recovery systems.

A robust neural network-based approach for microseismic event detection

KAUST Repository

Akram, Jubran; Ovcharenko, Oleg; Peter, Daniel

2017-01-01

We present an artificial neural network based approach for robust event detection from low S/N waveforms. We use a feed-forward network with a single hidden layer that is tuned on a training dataset and later applied on the entire example dataset

Biological engineering applications of feedforward neural networks designed and parameterized by genetic algorithms.

Science.gov (United States)

Ferentinos, Konstantinos P

2005-09-01

Two neural network (NN) applications in the field of biological engineering are developed, designed and parameterized by an evolutionary method based on the evolutionary process of genetic algorithms. The developed systems are a fault detection NN model and a predictive modeling NN system. An indirect or 'weak specification' representation was used for the encoding of NN topologies and training parameters into genes of the genetic algorithm (GA). Some a priori knowledge of the demands in network topology for specific application cases is required by this approach, so that the infinite search space of the problem is limited to some reasonable degree. Both one-hidden-layer and two-hidden-layer network architectures were explored by the GA. Except for the network architecture, each gene of the GA also encoded the type of activation functions in both hidden and output nodes of the NN and the type of minimization algorithm that was used by the backpropagation algorithm for the training of the NN. Both models achieved satisfactory performance, while the GA system proved to be a powerful tool that can successfully replace the problematic trial-and-error approach that is usually used for these tasks.

Multilayer Perceptron Neural Networks Model for Meteosat Second Generation SEVIRI Daytime Cloud Masking

Directory of Open Access Journals (Sweden)

Alireza Taravat

2015-02-01

Full Text Available A multilayer perceptron neural network cloud mask for Meteosat Second Generation SEVIRI (Spinning Enhanced Visible and Infrared Imager images is introduced and evaluated. The model is trained for cloud detection on MSG SEVIRI daytime data. It consists of a multi-layer perceptron with one hidden sigmoid layer, trained with the error back-propagation algorithm. The model is fed by six bands of MSG data (0.6, 0.8, 1.6, 3.9, 6.2 and 10.8 μm with 10 hidden nodes. The multiple-layer perceptrons lead to a cloud detection accuracy of 88.96%, when trained to map two predefined values that classify cloud and clear sky. The network was further evaluated using sixty MSG images taken at different dates. The network detected not only bright thick clouds but also thin or less bright clouds. The analysis demonstrated the feasibility of using machine learning models of cloud detection in MSG SEVIRI imagery.

Application of Artificial Neural Networks for estimating index floods

Science.gov (United States)

Šimor, Viliam; Hlavčová, Kamila; Kohnová, Silvia; Szolgay, Ján

2012-12-01

This article presents an application of Artificial Neural Networks (ANNs) and multiple regression models for estimating mean annual maximum discharge (index flood) at ungauged sites. Both approaches were tested for 145 small basins in Slovakia in areas ranging from 20 to 300 km2. Using the objective clustering method, the catchments were divided into ten homogeneous pooling groups; for each pooling group, mutually independent predictors (catchment characteristics) were selected for both models. The neural network was applied as a simple multilayer perceptron with one hidden layer and with a back propagation learning algorithm. Hyperbolic tangents were used as an activation function in the hidden layer. Estimating index floods by the multiple regression models were based on deriving relationships between the index floods and catchment predictors. The efficiencies of both approaches were tested by the Nash-Sutcliffe and a correlation coefficients. The results showed the comparative applicability of both models with slightly better results for the index floods achieved using the ANNs methodology.

Forecasting SPEI and SPI Drought Indices Using the Integrated Artificial Neural Networks

Directory of Open Access Journals (Sweden)

Petr Maca

2016-01-01

Full Text Available The presented paper compares forecast of drought indices based on two different models of artificial neural networks. The first model is based on feedforward multilayer perceptron, sANN, and the second one is the integrated neural network model, hANN. The analyzed drought indices are the standardized precipitation index (SPI and the standardized precipitation evaporation index (SPEI and were derived for the period of 1948–2002 on two US catchments. The meteorological and hydrological data were obtained from MOPEX experiment. The training of both neural network models was made by the adaptive version of differential evolution, JADE. The comparison of models was based on six model performance measures. The results of drought indices forecast, explained by the values of four model performance indices, show that the integrated neural network model was superior to the feedforward multilayer perceptron with one hidden layer of neurons.

Foreground removal from WMAP 5 yr temperature maps using an MLP neural network

DEFF Research Database (Denmark)

Nørgaard-Nielsen, Hans Ulrik

2010-01-01

CMB signal makes it essential to minimize the systematic errors in the CMB temperature determinations. Methods. The feasibility of using simple neural networks to extract the CMB signal from detailed simulated data has already been demonstrated. Here, simple neural networks are applied to the WMAP 5...... yr temperature data without using any auxiliary data. Results. A simple multilayer perceptron neural network with two hidden layers provides temperature estimates over more than 75 per cent of the sky with random errors significantly below those previously extracted from these data. Also......, the systematic errors, i.e. errors correlated with the Galactic foregrounds, are very small. Conclusions. With these results the neural network method is well prepared for dealing with the high-quality CMB data from the ESA Planck Surveyor satellite. © ESO, 2010....

Study on algorithm of process neural network for soft sensing in sewage disposal system

Science.gov (United States)

Liu, Zaiwen; Xue, Hong; Wang, Xiaoyi; Yang, Bin; Lu, Siying

2006-11-01

A new method of soft sensing based on process neural network (PNN) for sewage disposal system is represented in the paper. PNN is an extension of traditional neural network, in which the inputs and outputs are time-variation. An aggregation operator is introduced to process neuron, and it makes the neuron network has the ability to deal with the information of space-time two dimensions at the same time, so the data processing enginery of biological neuron is imitated better than traditional neuron. Process neural network with the structure of three layers in which hidden layer is process neuron and input and output are common neurons for soft sensing is discussed. The intelligent soft sensing based on PNN may be used to fulfill measurement of the effluent BOD (Biochemical Oxygen Demand) from sewage disposal system, and a good training result of soft sensing was obtained by the method.

Artificial Neural Networks to Detect Risk of Type 2 Diabetes | Baha ...

African Journals Online (AJOL)

A multilayer feedforward architecture with backpropagation algorithm was designed using Neural Network Toolbox of Matlab. The network was trained using batch mode backpropagation with gradient descent and momentum. Best performed network identified during the training was 2 hidden layers of 6 and 3 neurons, ...

Neural field model of memory-guided search.

Science.gov (United States)

Kilpatrick, Zachary P; Poll, Daniel B

2017-12-01

Many organisms can remember locations they have previously visited during a search. Visual search experiments have shown exploration is guided away from these locations, reducing redundancies in the search path before finding a hidden target. We develop and analyze a two-layer neural field model that encodes positional information during a search task. A position-encoding layer sustains a bump attractor corresponding to the searching agent's current location, and search is modeled by velocity input that propagates the bump. A memory layer sustains persistent activity bounded by a wave front, whose edges expand in response to excitatory input from the position layer. Search can then be biased in response to remembered locations, influencing velocity inputs to the position layer. Asymptotic techniques are used to reduce the dynamics of our model to a low-dimensional system of equations that track the bump position and front boundary. Performance is compared for different target-finding tasks.

Neural field model of memory-guided search

Science.gov (United States)

Kilpatrick, Zachary P.; Poll, Daniel B.

2017-12-01

Many organisms can remember locations they have previously visited during a search. Visual search experiments have shown exploration is guided away from these locations, reducing redundancies in the search path before finding a hidden target. We develop and analyze a two-layer neural field model that encodes positional information during a search task. A position-encoding layer sustains a bump attractor corresponding to the searching agent's current location, and search is modeled by velocity input that propagates the bump. A memory layer sustains persistent activity bounded by a wave front, whose edges expand in response to excitatory input from the position layer. Search can then be biased in response to remembered locations, influencing velocity inputs to the position layer. Asymptotic techniques are used to reduce the dynamics of our model to a low-dimensional system of equations that track the bump position and front boundary. Performance is compared for different target-finding tasks.

Biased Dropout and Crossmap Dropout: Learning towards effective Dropout regularization in convolutional neural network.

Science.gov (United States)

Poernomo, Alvin; Kang, Dae-Ki

2018-08-01

Training a deep neural network with a large number of parameters often leads to overfitting problem. Recently, Dropout has been introduced as a simple, yet effective regularization approach to combat overfitting in such models. Although Dropout has shown remarkable results on many deep neural network cases, its actual effect on CNN has not been thoroughly explored. Moreover, training a Dropout model will significantly increase the training time as it takes longer time to converge than a non-Dropout model with the same architecture. To deal with these issues, we address Biased Dropout and Crossmap Dropout, two novel approaches of Dropout extension based on the behavior of hidden units in CNN model. Biased Dropout divides the hidden units in a certain layer into two groups based on their magnitude and applies different Dropout rate to each group appropriately. Hidden units with higher activation value, which give more contributions to the network final performance, will be retained by a lower Dropout rate, while units with lower activation value will be exposed to a higher Dropout rate to compensate the previous part. The second approach is Crossmap Dropout, which is an extension of the regular Dropout in convolution layer. Each feature map in a convolution layer has a strong correlation between each other, particularly in every identical pixel location in each feature map. Crossmap Dropout tries to maintain this important correlation yet at the same time break the correlation between each adjacent pixel with respect to all feature maps by applying the same Dropout mask to all feature maps, so that all pixels or units in equivalent positions in each feature map will be either dropped or active during training. Our experiment with various benchmark datasets shows that our approaches provide better generalization than the regular Dropout. Moreover, our Biased Dropout takes faster time to converge during training phase, suggesting that assigning noise appropriately in

Task-specific feature extraction and classification of fMRI volumes using a deep neural network initialized with a deep belief network: Evaluation using sensorimotor tasks.

Science.gov (United States)

Jang, Hojin; Plis, Sergey M; Calhoun, Vince D; Lee, Jong-Hwan

2017-01-15

Feedforward deep neural networks (DNNs), artificial neural networks with multiple hidden layers, have recently demonstrated a record-breaking performance in multiple areas of applications in computer vision and speech processing. Following the success, DNNs have been applied to neuroimaging modalities including functional/structural magnetic resonance imaging (MRI) and positron-emission tomography data. However, no study has explicitly applied DNNs to 3D whole-brain fMRI volumes and thereby extracted hidden volumetric representations of fMRI that are discriminative for a task performed as the fMRI volume was acquired. Our study applied fully connected feedforward DNN to fMRI volumes collected in four sensorimotor tasks (i.e., left-hand clenching, right-hand clenching, auditory attention, and visual stimulus) undertaken by 12 healthy participants. Using a leave-one-subject-out cross-validation scheme, a restricted Boltzmann machine-based deep belief network was pretrained and used to initialize weights of the DNN. The pretrained DNN was fine-tuned while systematically controlling weight-sparsity levels across hidden layers. Optimal weight-sparsity levels were determined from a minimum validation error rate of fMRI volume classification. Minimum error rates (mean±standard deviation; %) of 6.9 (±3.8) were obtained from the three-layer DNN with the sparsest condition of weights across the three hidden layers. These error rates were even lower than the error rates from the single-layer network (9.4±4.6) and the two-layer network (7.4±4.1). The estimated DNN weights showed spatial patterns that are remarkably task-specific, particularly in the higher layers. The output values of the third hidden layer represented distinct patterns/codes of the 3D whole-brain fMRI volume and encoded the information of the tasks as evaluated from representational similarity analysis. Our reported findings show the ability of the DNN to classify a single fMRI volume based on the

Internal measuring models in trained neural networks for parameter estimation from images

NARCIS (Netherlands)

Feng, Tian-Jin; Feng, T.J.; Houkes, Z.; Korsten, Maarten J.; Spreeuwers, Lieuwe Jan

1992-01-01

The internal representations of 'learned' knowledge in neural networks are still poorly understood, even for backpropagation networks. The paper discusses a possible interpretation of learned knowledge of a network trained for parameter estimation from images. The outputs of the hidden layer are the

Predicting carbonate permeabilities from wireline logs using a back-propagation neural network

International Nuclear Information System (INIS)

Wiener, J.M.; Moll, R.F.; Rogers, J.A.

1991-01-01

This paper explores the applicability of using Neural Networks to aid in the determination of carbonate permeability from wireline logs. Resistivity, interval transit time, neutron porosity, and bulk density logs form Texaco's Stockyard Creek Oil field were used as input to a specially designed neural network to predict core permeabilities in this carbonate reservoir. Also of interest was the comparison of the neural network's results to those of standard statistical techniques. The process of developing the neural network for this problem has shown that a good understanding of the data is required when creating the training set from which the network learns. This network was trained to learn core permeabilities from raw and transformed log data using a hyperbolic tangent transfer function and a sum of squares global error function. Also, it required two hidden layers to solve this particular problem

A Simple Quantum Neural Net with a Periodic Activation Function

OpenAIRE

Daskin, Ammar

2018-01-01

In this paper, we propose a simple neural net that requires only $O(nlog_2k)$ number of qubits and $O(nk)$ quantum gates: Here, $n$ is the number of input parameters, and $k$ is the number of weights applied to these parameters in the proposed neural net. We describe the network in terms of a quantum circuit, and then draw its equivalent classical neural net which involves $O(k^n)$ nodes in the hidden layer. Then, we show that the network uses a periodic activation function of cosine values o...

Development and application of deep convolutional neural network in target detection

Science.gov (United States)

Jiang, Xiaowei; Wang, Chunping; Fu, Qiang

2018-04-01

With the development of big data and algorithms, deep convolution neural networks with more hidden layers have more powerful feature learning and feature expression ability than traditional machine learning methods, making artificial intelligence surpass human level in many fields. This paper first reviews the development and application of deep convolutional neural networks in the field of object detection in recent years, then briefly summarizes and ponders some existing problems in the current research, and the future development of deep convolutional neural network is prospected.

An artificial neural network model for periodic trajectory generation

Science.gov (United States)

Shankar, S.; Gander, R. E.; Wood, H. C.

A neural network model based on biological systems was developed for potential robotic application. The model consists of three interconnected layers of artificial neurons or units: an input layer subdivided into state and plan units, an output layer, and a hidden layer between the two outer layers which serves to implement nonlinear mappings between the input and output activation vectors. Weighted connections are created between the three layers, and learning is effected by modifying these weights. Feedback connections between the output and the input state serve to make the network operate as a finite state machine. The activation vector of the plan units of the input layer emulates the supraspinal commands in biological central pattern generators in that different plan activation vectors correspond to different sequences or trajectories being recalled, even with different frequencies. Three trajectories were chosen for implementation, and learning was accomplished in 10,000 trials. The fault tolerant behavior, adaptiveness, and phase maintenance of the implemented network are discussed.

Classification of E-Nose Aroma Data of Four Fruit Types by ABC-Based Neural Network

Directory of Open Access Journals (Sweden)

M. Fatih Adak

2016-02-01

Full Text Available Electronic nose technology is used in many areas, and frequently in the beverage industry for classification and quality-control purposes. In this study, four different aroma data (strawberry, lemon, cherry, and melon were obtained using a MOSES II electronic nose for the purpose of fruit classification. To improve the performance of the classification, the training phase of the neural network with two hidden layers was optimized using artificial bee colony algorithm (ABC, which is known to be successful in exploration. Test data were given to two different neural networks, each of which were trained separately with backpropagation (BP and ABC, and average test performances were measured as 60% for the artificial neural network trained with BP and 76.39% for the artificial neural network trained with ABC. Training and test phases were repeated 30 times to obtain these average performance measurements. This level of performance shows that the artificial neural network trained with ABC is successful in classifying aroma data.

Classification of E-Nose Aroma Data of Four Fruit Types by ABC-Based Neural Network.

Science.gov (United States)

Adak, M Fatih; Yumusak, Nejat

2016-02-27

Electronic nose technology is used in many areas, and frequently in the beverage industry for classification and quality-control purposes. In this study, four different aroma data (strawberry, lemon, cherry, and melon) were obtained using a MOSES II electronic nose for the purpose of fruit classification. To improve the performance of the classification, the training phase of the neural network with two hidden layers was optimized using artificial bee colony algorithm (ABC), which is known to be successful in exploration. Test data were given to two different neural networks, each of which were trained separately with backpropagation (BP) and ABC, and average test performances were measured as 60% for the artificial neural network trained with BP and 76.39% for the artificial neural network trained with ABC. Training and test phases were repeated 30 times to obtain these average performance measurements. This level of performance shows that the artificial neural network trained with ABC is successful in classifying aroma data.

Exploiting Hidden Layer Responses of Deep Neural Networks for Language Recognition

Science.gov (United States)

2016-09-08

Target Languages Arabic (ara) Egyptian , Iraqi, Levantine, Maghrebi,Modern Standard Chinese (chi) Cantonese, Mandarin, Min, Wu English (eng) British...Frame-by-frame DNN classification x1 x2 x3 xT-­1xT Figure 1: Frame-by-frame DNN Language Identification Figure 1 shows the architecture of the DNN...compare direct DNN system with proposed DNN I-vector system, we trained a single neural network to classify all 20 languages. The architecture of this

Application of RBF neural network improved by peak density function in intelligent color matching of wood dyeing

International Nuclear Information System (INIS)

Guan, Xuemei; Zhu, Yuren; Song, Wenlong

2016-01-01

According to the characteristics of wood dyeing, we propose a predictive model of pigment formula for wood dyeing based on Radial Basis Function (RBF) neural network. In practical application, however, it is found that the number of neurons in the hidden layer of RBF neural network is difficult to determine. In general, we need to test several times according to experience and prior knowledge, which is lack of a strict design procedure on theoretical basis. And we also don’t know whether the RBF neural network is convergent. This paper proposes a peak density function to determine the number of neurons in the hidden layer. In contrast to existing approaches, the centers and the widths of the radial basis function are initialized by extracting the features of samples. So the uncertainty caused by random number when initializing the training parameters and the topology of RBF neural network is eliminated. The average relative error of the original RBF neural network is 1.55% in 158 epochs. However, the average relative error of the RBF neural network which is improved by peak density function is only 0.62% in 50 epochs. Therefore, the convergence rate and approximation precision of the RBF neural network are improved significantly.

A neural network based computational model to predict the output power of different types of photovoltaic cells.

Science.gov (United States)

Xiao, WenBo; Nazario, Gina; Wu, HuaMing; Zhang, HuaMing; Cheng, Feng

2017-01-01

In this article, we introduced an artificial neural network (ANN) based computational model to predict the output power of three types of photovoltaic cells, mono-crystalline (mono-), multi-crystalline (multi-), and amorphous (amor-) crystalline. The prediction results are very close to the experimental data, and were also influenced by numbers of hidden neurons. The order of the solar generation power output influenced by the external conditions from smallest to biggest is: multi-, mono-, and amor- crystalline silicon cells. In addition, the dependences of power prediction on the number of hidden neurons were studied. For multi- and amorphous crystalline cell, three or four hidden layer units resulted in the high correlation coefficient and low MSEs. For mono-crystalline cell, the best results were achieved at the hidden layer unit of 8.

A neural network based computational model to predict the output power of different types of photovoltaic cells.

Directory of Open Access Journals (Sweden)

WenBo Xiao

Full Text Available In this article, we introduced an artificial neural network (ANN based computational model to predict the output power of three types of photovoltaic cells, mono-crystalline (mono-, multi-crystalline (multi-, and amorphous (amor- crystalline. The prediction results are very close to the experimental data, and were also influenced by numbers of hidden neurons. The order of the solar generation power output influenced by the external conditions from smallest to biggest is: multi-, mono-, and amor- crystalline silicon cells. In addition, the dependences of power prediction on the number of hidden neurons were studied. For multi- and amorphous crystalline cell, three or four hidden layer units resulted in the high correlation coefficient and low MSEs. For mono-crystalline cell, the best results were achieved at the hidden layer unit of 8.

A one-layer recurrent neural network for constrained nonsmooth invex optimization.

Science.gov (United States)

Li, Guocheng; Yan, Zheng; Wang, Jun

2014-02-01

Invexity is an important notion in nonconvex optimization. In this paper, a one-layer recurrent neural network is proposed for solving constrained nonsmooth invex optimization problems, designed based on an exact penalty function method. It is proved herein that any state of the proposed neural network is globally convergent to the optimal solution set of constrained invex optimization problems, with a sufficiently large penalty parameter. In addition, any neural state is globally convergent to the unique optimal solution, provided that the objective function and constraint functions are pseudoconvex. Moreover, any neural state is globally convergent to the feasible region in finite time and stays there thereafter. The lower bounds of the penalty parameter and convergence time are also estimated. Two numerical examples are provided to illustrate the performances of the proposed neural network. Copyright © 2013 Elsevier Ltd. All rights reserved.

CONSTRUCTION COST PREDICTION USING NEURAL NETWORKS

Directory of Open Access Journals (Sweden)

Smita K Magdum

2017-10-01

Full Text Available Construction cost prediction is important for construction firms to compete and grow in the industry. Accurate construction cost prediction in the early stage of project is important for project feasibility studies and successful completion. There are many factors that affect the cost prediction. This paper presents construction cost prediction as multiple regression model with cost of six materials as independent variables. The objective of this paper is to develop neural networks and multilayer perceptron based model for construction cost prediction. Different models of NN and MLP are developed with varying hidden layer size and hidden nodes. Four artificial neural network models and twelve multilayer perceptron models are compared. MLP and NN give better results than statistical regression method. As compared to NN, MLP works better on training dataset but fails on testing dataset. Five activation functions are tested to identify suitable function for the problem. ‘elu' transfer function gives better results than other transfer function.

Modeling the Constitutive Relationship of Al–0.62Mg–0.73Si Alloy Based on Artificial Neural Network

Directory of Open Access Journals (Sweden)

Ying Han

2017-03-01

Full Text Available In this work, the hot deformation behavior of 6A02 aluminum alloy was investigated by isothermal compression tests conducted in the temperature range of 683–783 K and strain-rate range of 0.001–1 s−1. According to the obtained true stress–true strain curves, the constitutive relationship of the alloy was revealed by establishing the Arrhenius-type constitutive model and back-propagation (BP neural network model. It is found that the flow characteristic of 6A02 aluminum alloy is closely related to deformation temperature and strain rate, and the true stress decreases with increasing temperatures and decreasing strain rates. The hot deformation activation energy is calculated to be 168.916 kJ mol−1. The BP neural network model with one hidden layer and 20 neurons in the hidden layer is developed. The accuracy in prediction of the Arrhenius-type constitutive model and BP neural network model is eveluated by using statistics analysis method. It is demonstrated that the BP neural network model has better performance in predicting the flow stress.

Genetic algorithms used to optimize an artificial neural network design used in neutron spectrometry

International Nuclear Information System (INIS)

Arteaga A, T.; Ortiz R, J. M.; Vega C, H. R.

2016-10-01

Artificial neural networks (Ann) are widely used; it which consist of an input layer, one or more hidden layers and an output layer; these layers contain neurons and each has connections called weights, where the knowledge are allowed and let to Ann solve problems proposed. These Ann is used to reconstruction of the energy spectrum of neutrons from count rates and develop Bonner sphere neutron dosimetry. Currently, we have developed Ann with high performance and generalization ability. Determine your optimal architecture is usually a difficult task, an exhaustive search of all possible combinations of parameters is rarely possible further training of the neural network with random initial weights can cause two major drawbacks: it can stuck in local minima or converge very slowly. In this project it will be used Genetic Algorithms (Ga); which are based on the principle or analogy of evolution through natural selection and has shown to be very effective in optimizing complex search functions and large spaces or to find a near optimal overall efficiency. The aim is to decrease the architecture in number of hidden neurons and therefore the total number of connections is reducing. The benefits obtained by optimizing the network are that the number of connections would be considerably smaller and thus the computational complexity, hardware integration, resources will be lower such that will allow to be even more viable implemented. To use the Ga three problems must be solve: 1) coding the problem into chromosomes. 2) Construct a fitness function. 3) Proper selection of genetic operators; crossover, selection, mutation. As a result, the scientific knowledge obtained can to be applied to similar problems having a reference parameters used and their impact on the optimization would to be generated. It concluded that the input layer and output are subject to the problem; the Ga propose the optimal number of neurons in the hidden layer without losing the quality of the

Artificial neural network based modelling approach for municipal solid waste gasification in a fluidized bed reactor.

Science.gov (United States)

Pandey, Daya Shankar; Das, Saptarshi; Pan, Indranil; Leahy, James J; Kwapinski, Witold

2016-12-01

In this paper, multi-layer feed forward neural networks are used to predict the lower heating value of gas (LHV), lower heating value of gasification products including tars and entrained char (LHV p ) and syngas yield during gasification of municipal solid waste (MSW) during gasification in a fluidized bed reactor. These artificial neural networks (ANNs) with different architectures are trained using the Levenberg-Marquardt (LM) back-propagation algorithm and a cross validation is also performed to ensure that the results generalise to other unseen datasets. A rigorous study is carried out on optimally choosing the number of hidden layers, number of neurons in the hidden layer and activation function in a network using multiple Monte Carlo runs. Nine input and three output parameters are used to train and test various neural network architectures in both multiple output and single output prediction paradigms using the available experimental datasets. The model selection procedure is carried out to ascertain the best network architecture in terms of predictive accuracy. The simulation results show that the ANN based methodology is a viable alternative which can be used to predict the performance of a fluidized bed gasifier. Copyright © 2016 Elsevier Ltd. All rights reserved.

Neural control of magnetic suspension systems

Science.gov (United States)

Gray, W. Steven

1993-01-01

The purpose of this research program is to design, build and test (in cooperation with NASA personnel from the NASA Langley Research Center) neural controllers for two different small air-gap magnetic suspension systems. The general objective of the program is to study neural network architectures for the purpose of control in an experimental setting and to demonstrate the feasibility of the concept. The specific objectives of the research program are: (1) to demonstrate through simulation and experimentation the feasibility of using neural controllers to stabilize a nonlinear magnetic suspension system; (2) to investigate through simulation and experimentation the performance of neural controllers designs under various types of parametric and nonparametric uncertainty; (3) to investigate through simulation and experimentation various types of neural architectures for real-time control with respect to performance and complexity; and (4) to benchmark in an experimental setting the performance of neural controllers against other types of existing linear and nonlinear compensator designs. To date, the first one-dimensional, small air-gap magnetic suspension system has been built, tested and delivered to the NASA Langley Research Center. The device is currently being stabilized with a digital linear phase-lead controller. The neural controller hardware is under construction. Two different neural network paradigms are under consideration, one based on hidden layer feedforward networks trained via back propagation and one based on using Gaussian radial basis functions trained by analytical methods related to stability conditions. Some advanced nonlinear control algorithms using feedback linearization and sliding mode control are in simulation studies.

Predicting the topology of dynamic neural networks for the simulation of electronic circuits

NARCIS (Netherlands)

Schilders, W.H.A.

2009-01-01

In this paper we discuss the use of the state-space modelling MOESP algorithm to generate precise information about the number of neurons and hidden layers in dynamic neural networks developed for the behavioural modelling of electronic circuits. The Bartels–Stewart algorithm is used to transform

Design of Jetty Piles Using Artificial Neural Networks

Directory of Open Access Journals (Sweden)

Yongjei Lee

2014-01-01

Full Text Available To overcome the complication of jetty pile design process, artificial neural networks (ANN are adopted. To generate the training samples for training ANN, finite element (FE analysis was performed 50 times for 50 different design cases. The trained ANN was verified with another FE analysis case and then used as a structural analyzer. The multilayer neural network (MBPNN with two hidden layers was used for ANN. The framework of MBPNN was defined as the input with the lateral forces on the jetty structure and the type of piles and the output with the stress ratio of the piles. The results from the MBPNN agree well with those from FE analysis. Particularly for more complex modes with hundreds of different design cases, the MBPNN would possibly substitute parametric studies with FE analysis saving design time and cost.

Prediction of Polymer Flooding Performance with an Artificial Neural Network: A Two-Polymer-Slug Case

Directory of Open Access Journals (Sweden)

Jestril Ebaga-Ololo

2017-07-01

Full Text Available Many previous contributions to methods of forecasting the performance of polymer flooding using artificial neural networks (ANNs have been made by numerous researchers previously. In most of those forecasting cases, only a single polymer slug was employed to meet the objective of the study. The intent of this manuscript is to propose an efficient recovery factor prediction tool at different injection stages of two polymer slugs during polymer flooding using an ANN. In this regard, a back-propagation algorithm was coupled with six input parameters to predict three output parameters via a hidden layer composed of 10 neurons. Evaluation of the ANN model performance was made with multiple linear regression. With an acceptable correlation coefficient, the proposed ANN tool was able to predict the recovery factor with errors of <1%. In addition, to understand the influence of each parameter on the output parameters, a sensitivity analysis was applied to the input parameters. The results showed less impact from the second polymer concentration, owing to changes in permeability after the injection of the first polymer slug.

[Rapid Identification of Epicarpium Citri Grandis via Infrared Spectroscopy and Fluorescence Spectrum Imaging Technology Combined with Neural Network].

Science.gov (United States)

Pan, Sha-sha; Huang, Fu-rong; Xiao, Chi; Xian, Rui-yi; Ma, Zhi-guo

2015-10-01

To explore rapid reliable methods for detection of Epicarpium citri grandis (ECG), the experiment using Fourier Transform Attenuated Total Reflection Infrared Spectroscopy (FTIR/ATR) and Fluorescence Spectrum Imaging Technology combined with Multilayer Perceptron (MLP) Neural Network pattern recognition, for the identification of ECG, and the two methods are compared. Infrared spectra and fluorescence spectral images of 118 samples, 81 ECG and 37 other kinds of ECG, are collected. According to the differences in tspectrum, the spectra data in the 550-1 800 cm(-1) wavenumber range and 400-720 nm wavelength are regarded as the study objects of discriminant analysis. Then principal component analysis (PCA) is applied to reduce the dimension of spectroscopic data of ECG and MLP Neural Network is used in combination to classify them. During the experiment were compared the effects of different methods of data preprocessing on the model: multiplicative scatter correction (MSC), standard normal variable correction (SNV), first-order derivative(FD), second-order derivative(SD) and Savitzky-Golay (SG). The results showed that: after the infrared spectra data via the Savitzky-Golay (SG) pretreatment through the MLP Neural Network with the hidden layer function as sigmoid, we can get the best discrimination of ECG, the correct percent of training set and testing set are both 100%. Using fluorescence spectral imaging technology, corrected by the multiple scattering (MSC) results in the pretreatment is the most ideal. After data preprocessing, the three layers of the MLP Neural Network of the hidden layer function as sigmoid function can get 100% correct percent of training set and 96.7% correct percent of testing set. It was shown that the FTIR/ATR and fluorescent spectral imaging technology combined with MLP Neural Network can be used for the identification study of ECG and has the advantages of rapid, reliable effect.

One-dimensional model of cable-in-conduit superconductors under cyclic loading using artificial neural networks

International Nuclear Information System (INIS)

Lefik, M.; Schrefler, B.A.

2002-01-01

An artificial neural network with two hidden layers is trained to define a mechanical constitutive relation for superconducting cable under transverse cyclic loading. The training is performed using a set of experimental data. The behaviour of the cable is strongly non-linear. Irreversible phenomena result with complicated loops of hysteresis. The performance of the ANN, which is applied as a tool for storage, interpolation and interpretation of experimental data is investigated, both from numerical, as well as from physical viewpoints

Behavioural modelling using the MOESP algorithm, dynamic neural networks and the Bartels-Stewart algorithm

NARCIS (Netherlands)

Schilders, W.H.A.; Meijer, P.B.L.; Ciggaar, E.

2008-01-01

In this paper we discuss the use of the state-space modelling MOESP algorithm to generate precise information about the number of neurons and hidden layers in dynamic neural networks developed for the behavioural modelling of electronic circuits. The Bartels–Stewart algorithm is used to transform

Development of Artificial Neural Network Model for Diesel Fuel Properties Prediction using Vibrational Spectroscopy.

Science.gov (United States)

Bolanča, Tomislav; Marinović, Slavica; Ukić, Sime; Jukić, Ante; Rukavina, Vinko

2012-06-01

This paper describes development of artificial neural network models which can be used to correlate and predict diesel fuel properties from several FTIR-ATR absorbances and Raman intensities as input variables. Multilayer feed forward and radial basis function neural networks have been used to rapid and simultaneous prediction of cetane number, cetane index, density, viscosity, distillation temperatures at 10% (T10), 50% (T50) and 90% (T90) recovery, contents of total aromatics and polycyclic aromatic hydrocarbons of commercial diesel fuels. In this study two-phase training procedures for multilayer feed forward networks were applied. While first phase training algorithm was constantly the back propagation one, two second phase training algorithms were varied and compared, namely: conjugate gradient and quasi Newton. In case of radial basis function network, radial layer was trained using K-means radial assignment algorithm and three different radial spread algorithms: explicit, isotropic and K-nearest neighbour. The number of hidden layer neurons and experimental data points used for the training set have been optimized for both neural networks in order to insure good predictive ability by reducing unnecessary experimental work. This work shows that developed artificial neural network models can determine main properties of diesel fuels simultaneously based on a single and fast IR or Raman measurement.

Storage capacity of multi-layered neural networks with binary weights

International Nuclear Information System (INIS)

Tarkowski, W.; Hemmen, J.L. van

1997-01-01

Using statistical physics methods we investigate two-layered perceptrons which consist of N binary input neurons, K hidden units and a single output node. Four basic types of such networks are considered: the so-called Committee, Parity, and AND Machines which makes a decision based on a majority, parity, and the logical AND rules, respectively (for these cases the weights that connect hidden units and output node are taken to be equal to one), and the General Machine where one allows all the synaptic couplings to vary. For these kinds of network we examine two types of architecture: fully connected and three-connected ones (with overlapping and non-overlapping receptive fields, respectively). All the above mentioned machines heave binary weights. Our basic interest is focused on the storage capabilities of such networks which realize p= αN random, unbiased dichotomies (α denotes the so-called storage ratio). The analysis is done using the annealed approximation and is valid for all values of K. The critical (maximal) storage capacity of the fully connected Committee Machine reads α c =K, while in the case of the three-structure one gets α c =1, independent of K. The results obtained for the Parity Machine are exactly the same as those for the Committee network. The optimal storage of the AND Machine depends on distribution of the outputs for the patterns. These associations are studied in detail. We have found also that the capacity of the General Machines remains the same as compared to systems with fixed weights between intermediate layer and the output node. Some of the findings (especially those concerning the storage capacity of the Parity Machine) are in a good agreement with known numerical results. (author)

Neural net classification of x-ray pistachio nut data

Science.gov (United States)

Casasent, David P.; Sipe, Michael A.; Schatzki, Thomas F.; Keagy, Pamela M.; Le, Lan Chau

1996-12-01

Classification results for agricultural products are presented using a new neural network. This neural network inherently produces higher-order decision surfaces. It achieves this with fewer hidden layer neurons than other classifiers require. This gives better generalization. It uses new techniques to select the number of hidden layer neurons and adaptive algorithms that avoid other such ad hoc parameter selection problems; it allows selection of the best classifier parameters without the need to analyze the test set results. The agriculture case study considered is the inspection and classification of pistachio nuts using x- ray imagery. Present inspection techniques cannot provide good rejection of worm damaged nuts without rejecting too many good nuts. X-ray imagery has the potential to provide 100% inspection of such agricultural products in real time. Only preliminary results are presented, but these indicate the potential to reduce major defects to 2% of the crop with 1% of good nuts rejected. Future image processing techniques that should provide better features to improve performance and allow inspection of a larger variety of nuts are noted. These techniques and variations of them have uses in a number of other agricultural product inspection problems.

Parameter diagnostics of phases and phase transition learning by neural networks

Science.gov (United States)

Suchsland, Philippe; Wessel, Stefan

2018-05-01

We present an analysis of neural network-based machine learning schemes for phases and phase transitions in theoretical condensed matter research, focusing on neural networks with a single hidden layer. Such shallow neural networks were previously found to be efficient in classifying phases and locating phase transitions of various basic model systems. In order to rationalize the emergence of the classification process and for identifying any underlying physical quantities, it is feasible to examine the weight matrices and the convolutional filter kernels that result from the learning process of such shallow networks. Furthermore, we demonstrate how the learning-by-confusing scheme can be used, in combination with a simple threshold-value classification method, to diagnose the learning parameters of neural networks. In particular, we study the classification process of both fully-connected and convolutional neural networks for the two-dimensional Ising model with extended domain wall configurations included in the low-temperature regime. Moreover, we consider the two-dimensional XY model and contrast the performance of the learning-by-confusing scheme and convolutional neural networks trained on bare spin configurations to the case of preprocessed samples with respect to vortex configurations. We discuss these findings in relation to similar recent investigations and possible further applications.

Application of improved PSO-RBF neural network in the synthetic ammonia decarbonization

Directory of Open Access Journals (Sweden)

Yongwei LI

2017-12-01

Full Text Available The synthetic ammonia decarbonization is a typical complex industrial process, which has the characteristics of time variation, nonlinearity and uncertainty, and the on-line control model is difficult to be established. An improved PSO-RBF neural network control algorithm is proposed to solve the problems of low precision and poor robustness in the complex process of the synthetic ammonia decarbonization. The particle swarm optimization algorithm and RBF neural network are combined. The improved particle swarm algorithm is used to optimize the RBF neural network's hidden layer primary function center, width and the output layer's connection value to construct the RBF neural network model optimized by the improved PSO algorithm. The improved PSO-RBF neural network control model is applied to the key carbonization process and compared with the traditional fuzzy neural network. The simulation results show that the improved PSO-RBF neural network control method used in the synthetic ammonia decarbonization process has higher control accuracy and system robustness, which provides an effective way to solve the modeling and optimization control of a complex industrial process.

Desien, ConstruThe design, fabrication and evaluation of egg weighing device using capacitive sensor and neural networksction and Evaluation of Egg Weighing Device Using Capacitive Sensor and Neural Networks

Directory of Open Access Journals (Sweden)

S Khalili

2015-09-01

egg-laying day, and the second and fourth day after laying. Results and Discussion: In this study, two networks were built and evaluated. In the first series, two-layer networks and in the second series, three-layer networks were developed. In the two-layer neural networks, the number of neurons in the hidden layer was changed from 2 to 10.According to the given results for two-layer networks, two layer networks with 10 neurons offer the best results (the highest R-value and minimum RMSE and it can be chosen as the most effective two-layer network. Three-layer neural networks have been composed of two hidden layers. The number of neurons in the first hidden layer was 10 and in the second layer it was changed from 1 to 20. Between three-layer networks, the network with 7 neurons with the highest R-value and the lowest error is the most appropriate network. It is even more efficient than the two-layer network with 10 neurons. So, the most appropriate structure is 1-7-10-16 and it has been selected for calibration of the weighing device. To evaluate and assess the accuracy of the weighing machine, weights of 24 samples of fresh eggs were predicted and compared with the actual values obtained using a digital scale with the accuracy of 0.01 gr. The paired t-test has been used to compare the measured and predicted values and the Bland-Altman method has been used for charting the accordance between the measured and predicted values. Based on the findings, the difference between the measured and predicted values was observed up to 5.4 gr that is related to a very large sample. The mean absolute error is equal to 2.21 gr and the mean absolute percentage error is equal to 3.75 %. According to the findings, 95% of the actual and approximate matching range to compare the two weighing methods is between -5.3 gr and 3.36 gr. Thus, the dielectric technique may underestimate the egg weight up to 5.3 gr or it may overestimate it up to 3.36 gr more than the actual prediction

A clinical decision support system using multilayer perceptron neural network to assess well being in diabetes.

Science.gov (United States)

Narasingarao, M R; Manda, R; Sridhar, G R; Madhu, K; Rao, A A

2009-02-01

is 1 and the number of units in the hidden layer are 6, the normalized system error was 470.57. With input samples of 100, 150 and 200, keeping the other variables constant, the normalized system error was 419.61, 359.67 and 332.32 respectively. Similar values are found for the normalized system error when the number of units in the hidden layer have been increased to 7, 8 and 9 respectively. With two hidden layers, and with each hidden layer containing 6,7 ,8, 9, 10, 11 units for the samples 50, 100, 150, and 200, the same values of normalized system error was found. Women having weight between 40 kgs and 85kgs had higher levels of depression than men who had weight between 39kgs and 102 kgs. We have developed a prototype neural network model to predict the psychosocial well-being in diabetes, when biological or biographical variables are given as inputs. When greater data was fed to the system, the normalized system error can be reduced.

Spatial frequency domain spectroscopy of two layer media

Science.gov (United States)

Yudovsky, Dmitry; Durkin, Anthony J.

2011-10-01

Monitoring of tissue blood volume and oxygen saturation using biomedical optics techniques has the potential to inform the assessment of tissue health, healing, and dysfunction. These quantities are typically estimated from the contribution of oxyhemoglobin and deoxyhemoglobin to the absorption spectrum of the dermis. However, estimation of blood related absorption in superficial tissue such as the skin can be confounded by the strong absorption of melanin in the epidermis. Furthermore, epidermal thickness and pigmentation varies with anatomic location, race, gender, and degree of disease progression. This study describes a technique for decoupling the effect of melanin absorption in the epidermis from blood absorption in the dermis for a large range of skin types and thicknesses. An artificial neural network was used to map input optical properties to spatial frequency domain diffuse reflectance of two layer media. Then, iterative fitting was used to determine the optical properties from simulated spatial frequency domain diffuse reflectance. Additionally, an artificial neural network was trained to directly map spatial frequency domain reflectance to sets of optical properties of a two layer medium, thus bypassing the need for iteration. In both cases, the optical thickness of the epidermis and absorption and reduced scattering coefficients of the dermis were determined independently. The accuracy and efficiency of the iterative fitting approach was compared with the direct neural network inversion.

Foreground removal from CMB temperature maps using an MLP neural network

DEFF Research Database (Denmark)

Nørgaard-Nielsen, Hans Ulrik; Jørgensen, H.E.

2008-01-01

the CMB temperature signal from the combined signal CMB and the foregrounds has been investigated. As a specific example, we have analysed simulated data, as expected from the ESA Planck CMB mission. A simple multilayer perceptron neural network with 2 hidden layers can provide temperature estimates over...... CMB signal it is essential to minimize the systematic errors in the CMB temperature determinations. Following the available knowledge of the spectral behavior of the Galactic foregrounds simple power law-like spectra have been assumed. The feasibility of using a simple neural network for extracting...

Passivation and control of partially known SISO nonlinear systems via dynamic neural networks

Directory of Open Access Journals (Sweden)

Reyes-Reyes J.

2000-01-01

Full Text Available In this paper, an adaptive technique is suggested to provide the passivity property for a class of partially known SISO nonlinear systems. A simple Dynamic Neural Network (DNN, containing only two neurons and without any hidden-layers, is used to identify the unknown nonlinear system. By means of a Lyapunov-like analysis the new learning law for this DNN, guarantying both successful identification and passivation effects, is derived. Based on this adaptive DNN model, an adaptive feedback controller, serving for wide class of nonlinear systems with an a priori incomplete model description, is designed. Two typical examples illustrate the effectiveness of the suggested approach.

Evaluation of thermal conductivity of MgO-MWCNTs/EG hybrid nanofluids based on experimental data by selecting optimal artificial neural networks

Science.gov (United States)

Vafaei, Masoud; Afrand, Masoud; Sina, Nima; Kalbasi, Rasool; Sourani, Forough; Teimouri, Hamid

2017-01-01

In this paper, the thermal conductivity ratio of MgO-MWCNTs/EG hybrid nanofluids has been predicted by an optimal artificial neural network at solid volume fractions of 0.05%, 0.1%, 0.15%, 0.2%, 0.4% and 0.6% in the temperature range of 25-50 °C. In this way, at the first, thirty six experimental data was presented to determine the thermal conductivity ratio of the hybrid nanofluid. Then, four optimal artificial neural networks with 6, 8, 10 and 12 neurons in hidden layer were designed to predict the thermal conductivity ratio of the nanofluid. The comparison between four optimal ANN results and experimental showed that the ANN with 12 neurons in hidden layer was the best model. Moreover, the results obtained from the best ANN indicated the maximum deviation margin of 0.8%.

Three-dimensional sound localisation with a lizard peripheral auditory model

DEFF Research Database (Denmark)

Kjær Schmidt, Michael; Shaikh, Danish

the networks learned a transfer function that translated the three-dimensional non-linear mapping into estimated azimuth and elevation values for the acoustic target. The neural network with two hidden layers as expected performed better than that with only one hidden layer. Our approach assumes that for any...... location of an acoustic target in three dimensions. Our approach utilises a model of the peripheral auditory system of lizards [Christensen-Dalsgaard and Manley 2005] coupled with a multi-layer perceptron neural network. The peripheral auditory model’s response to sound input encodes sound direction...... information in a single plane which by itself is insufficient to localise the acoustic target in three dimensions. A multi-layer perceptron neural network is used to combine two independent responses of the model, corresponding to two rotational movements, into an estimate of the sound direction in terms...

[The Identification of the Origin of Chinese Wolfberry Based on Infrared Spectral Technology and the Artificial Neural Network].

Science.gov (United States)

Li, Zhong; Liu, Ming-de; Ji, Shou-xiang

2016-03-01

The Fourier Transform Infrared Spectroscopy (FTIR) is established to find the geographic origins of Chinese wolfberry quickly. In the paper, the 45 samples of Chinese wolfberry from different places of Qinghai Province are to be surveyed by FTIR. The original data matrix of FTIR is pretreated with common preprocessing and wavelet transform. Compared with common windows shifting smoothing preprocessing, standard normal variation correction and multiplicative scatter correction, wavelet transform is an effective spectrum data preprocessing method. Before establishing model through the artificial neural networks, the spectra variables are compressed by means of the wavelet transformation so as to enhance the training speed of the artificial neural networks, and at the same time the related parameters of the artificial neural networks model are also discussed in detail. The survey shows even if the infrared spectroscopy data is compressed to 1/8 of its original data, the spectral information and analytical accuracy are not deteriorated. The compressed spectra variables are used for modeling parameters of the backpropagation artificial neural network (BP-ANN) model and the geographic origins of Chinese wolfberry are used for parameters of export. Three layers of neural network model are built to predict the 10 unknown samples by using the MATLAB neural network toolbox design error back propagation network. The number of hidden layer neurons is 5, and the number of output layer neuron is 1. The transfer function of hidden layer is tansig, while the transfer function of output layer is purelin. Network training function is trainl and the learning function of weights and thresholds is learngdm. net. trainParam. epochs=1 000, while net. trainParam. goal = 0.001. The recognition rate of 100% is to be achieved. It can be concluded that the method is quite suitable for the quick discrimination of producing areas of Chinese wolfberry. The infrared spectral analysis technology

Prediction of geomagnetic storm using neural networks: Comparison of the efficiency of the Satellite and ground-based input parameters

International Nuclear Information System (INIS)

Stepanova, Marina; Antonova, Elizavieta; Munos-Uribe, F A; Gordo, S L Gomez; Torres-Sanchez, M V

2008-01-01

Different kinds of neural networks have established themselves as an effective tool in the prediction of different geomagnetic indices, including the Dst being the most important constituent for determination of the impact of Space Weather on the human life. Feed-forward networks with one hidden layer are used to forecast the Dst variation, using separately the solar wind paramenters, polar cap index, and auroral electrojet index as input parameters. It was found that in all three cases the storm-time intervals were predicted much more precisely as quite time intervals. The majority of cross-correlation coefficients between predicted and observed Dst of strong geomagnetic storms are situated between 0.8 and 0.9. Changes in the neural network architecture, including the number of nodes in the input and hidden layers and the transfer functions between them lead to an improvement of a network performance up to 10%.

Using deep recurrent neural network for direct beam solar irradiance cloud screening

Science.gov (United States)

Chen, Maosi; Davis, John M.; Liu, Chaoshun; Sun, Zhibin; Zempila, Melina Maria; Gao, Wei

2017-09-01

Cloud screening is an essential procedure for in-situ calibration and atmospheric properties retrieval on (UV-)MultiFilter Rotating Shadowband Radiometer [(UV-)MFRSR]. Previous study has explored a cloud screening algorithm for direct-beam (UV-)MFRSR voltage measurements based on the stability assumption on a long time period (typically a half day or a whole day). To design such an algorithm requires in-depth understanding of radiative transfer and delicate data manipulation. Recent rapid developments on deep neural network and computation hardware have opened a window for modeling complicated End-to-End systems with a standardized strategy. In this study, a multi-layer dynamic bidirectional recurrent neural network is built for determining the cloudiness on each time point with a 17-year training dataset and tested with another 1-year dataset. The dataset is the daily 3-minute cosine corrected voltages, airmasses, and the corresponding cloud/clear-sky labels at two stations of the USDA UV-B Monitoring and Research Program. The results show that the optimized neural network model (3-layer, 250 hidden units, and 80 epochs of training) has an overall test accuracy of 97.87% (97.56% for the Oklahoma site and 98.16% for the Hawaii site). Generally, the neural network model grasps the key concept of the original model to use data in the entire day rather than short nearby measurements to perform cloud screening. A scrutiny of the logits layer suggests that the neural network model automatically learns a way to calculate a quantity similar to total optical depth and finds an appropriate threshold for cloud screening.

Artificial neural network modeling of jatropha oil fueled diesel engine for emission predictions

Directory of Open Access Journals (Sweden)

Ganapathy Thirunavukkarasu

2009-01-01

Full Text Available This paper deals with artificial neural network modeling of diesel engine fueled with jatropha oil to predict the unburned hydrocarbons, smoke, and NOx emissions. The experimental data from the literature have been used as the data base for the proposed neural network model development. For training the networks, the injection timing, injector opening pressure, plunger diameter, and engine load are used as the input layer. The outputs are hydrocarbons, smoke, and NOx emissions. The feed forward back propagation learning algorithms with two hidden layers are used in the networks. For each output a different network is developed with required topology. The artificial neural network models for hydrocarbons, smoke, and NOx emissions gave R2 values of 0.9976, 0.9976, and 0.9984 and mean percent errors of smaller than 2.7603, 4.9524, and 3.1136, respectively, for training data sets, while the R2 values of 0.9904, 0.9904, and 0.9942, and mean percent errors of smaller than 6.5557, 6.1072, and 4.4682, respectively, for testing data sets. The best linear fit of regression to the artificial neural network models of hydrocarbons, smoke, and NOx emissions gave the correlation coefficient values of 0.98, 0.995, and 0.997, respectively.

Comparison between extreme learning machine and wavelet neural networks in data classification

Science.gov (United States)

Yahia, Siwar; Said, Salwa; Jemai, Olfa; Zaied, Mourad; Ben Amar, Chokri

2017-03-01

Extreme learning Machine is a well known learning algorithm in the field of machine learning. It's about a feed forward neural network with a single-hidden layer. It is an extremely fast learning algorithm with good generalization performance. In this paper, we aim to compare the Extreme learning Machine with wavelet neural networks, which is a very used algorithm. We have used six benchmark data sets to evaluate each technique. These datasets Including Wisconsin Breast Cancer, Glass Identification, Ionosphere, Pima Indians Diabetes, Wine Recognition and Iris Plant. Experimental results have shown that both extreme learning machine and wavelet neural networks have reached good results.

Diagnosing Autism Spectrum Disorder from Brain Resting-State Functional Connectivity Patterns Using a Deep Neural Network with a Novel Feature Selection Method.

Science.gov (United States)

Guo, Xinyu; Dominick, Kelli C; Minai, Ali A; Li, Hailong; Erickson, Craig A; Lu, Long J

2017-01-01

The whole-brain functional connectivity (FC) pattern obtained from resting-state functional magnetic resonance imaging data are commonly applied to study neuropsychiatric conditions such as autism spectrum disorder (ASD) by using different machine learning models. Recent studies indicate that both hyper- and hypo- aberrant ASD-associated FCs were widely distributed throughout the entire brain rather than only in some specific brain regions. Deep neural networks (DNN) with multiple hidden layers have shown the ability to systematically extract lower-to-higher level information from high dimensional data across a series of neural hidden layers, significantly improving classification accuracy for such data. In this study, a DNN with a novel feature selection method (DNN-FS) is developed for the high dimensional whole-brain resting-state FC pattern classification of ASD patients vs. typical development (TD) controls. The feature selection method is able to help the DNN generate low dimensional high-quality representations of the whole-brain FC patterns by selecting features with high discriminating power from multiple trained sparse auto-encoders. For the comparison, a DNN without the feature selection method (DNN-woFS) is developed, and both of them are tested with different architectures (i.e., with different numbers of hidden layers/nodes). Results show that the best classification accuracy of 86.36% is generated by the DNN-FS approach with 3 hidden layers and 150 hidden nodes (3/150). Remarkably, DNN-FS outperforms DNN-woFS for all architectures studied. The most significant accuracy improvement was 9.09% with the 3/150 architecture. The method also outperforms other feature selection methods, e.g., two sample t -test and elastic net. In addition to improving the classification accuracy, a Fisher's score-based biomarker identification method based on the DNN is also developed, and used to identify 32 FCs related to ASD. These FCs come from or cross different pre

Diagnosing Autism Spectrum Disorder from Brain Resting-State Functional Connectivity Patterns Using a Deep Neural Network with a Novel Feature Selection Method

Directory of Open Access Journals (Sweden)

Xinyu Guo

2017-08-01

Full Text Available The whole-brain functional connectivity (FC pattern obtained from resting-state functional magnetic resonance imaging data are commonly applied to study neuropsychiatric conditions such as autism spectrum disorder (ASD by using different machine learning models. Recent studies indicate that both hyper- and hypo- aberrant ASD-associated FCs were widely distributed throughout the entire brain rather than only in some specific brain regions. Deep neural networks (DNN with multiple hidden layers have shown the ability to systematically extract lower-to-higher level information from high dimensional data across a series of neural hidden layers, significantly improving classification accuracy for such data. In this study, a DNN with a novel feature selection method (DNN-FS is developed for the high dimensional whole-brain resting-state FC pattern classification of ASD patients vs. typical development (TD controls. The feature selection method is able to help the DNN generate low dimensional high-quality representations of the whole-brain FC patterns by selecting features with high discriminating power from multiple trained sparse auto-encoders. For the comparison, a DNN without the feature selection method (DNN-woFS is developed, and both of them are tested with different architectures (i.e., with different numbers of hidden layers/nodes. Results show that the best classification accuracy of 86.36% is generated by the DNN-FS approach with 3 hidden layers and 150 hidden nodes (3/150. Remarkably, DNN-FS outperforms DNN-woFS for all architectures studied. The most significant accuracy improvement was 9.09% with the 3/150 architecture. The method also outperforms other feature selection methods, e.g., two sample t-test and elastic net. In addition to improving the classification accuracy, a Fisher's score-based biomarker identification method based on the DNN is also developed, and used to identify 32 FCs related to ASD. These FCs come from or cross

Modeling of an industrial process of pleuromutilin fermentation using feed-forward neural networks

Directory of Open Access Journals (Sweden)

L. Khaouane

2013-03-01

Full Text Available This work investigates the use of artificial neural networks in modeling an industrial fermentation process of Pleuromutilin produced by Pleurotus mutilus in a fed-batch mode. Three feed-forward neural network models characterized by a similar structure (five neurons in the input layer, one hidden layer and one neuron in the output layer are constructed and optimized with the aim to predict the evolution of three main bioprocess variables: biomass, substrate and product. Results show a good fit between the predicted and experimental values for each model (the root mean squared errors were 0.4624% - 0.1234 g/L and 0.0016 mg/g respectively. Furthermore, the comparison between the optimized models and the unstructured kinetic models in terms of simulation results shows that neural network models gave more significant results. These results encourage further studies to integrate the mathematical formulae extracted from these models into an industrial control loop of the process.

Construction of Neural Networks for Realization of Localized Deep Learning

Directory of Open Access Journals (Sweden)

Charles K. Chui

2018-05-01

Full Text Available The subject of deep learning has recently attracted users of machine learning from various disciplines, including: medical diagnosis and bioinformatics, financial market analysis and online advertisement, speech and handwriting recognition, computer vision and natural language processing, time series forecasting, and search engines. However, theoretical development of deep learning is still at its infancy. The objective of this paper is to introduce a deep neural network (also called deep-net approach to localized manifold learning, with each hidden layer endowed with a specific learning task. For the purpose of illustrations, we only focus on deep-nets with three hidden layers, with the first layer for dimensionality reduction, the second layer for bias reduction, and the third layer for variance reduction. A feedback component is also designed to deal with outliers. The main theoretical result in this paper is the order O(m-2s/(2s+d of approximation of the regression function with regularity s, in terms of the number m of sample points, where the (unknown manifold dimension d replaces the dimension D of the sampling (Euclidean space for shallow nets.

Learning text representation using recurrent convolutional neural network with highway layers

OpenAIRE

Wen, Ying; Zhang, Weinan; Luo, Rui; Wang, Jun

2016-01-01

Recently, the rapid development of word embedding and neural networks has brought new inspiration to various NLP and IR tasks. In this paper, we describe a staged hybrid model combining Recurrent Convolutional Neural Networks (RCNN) with highway layers. The highway network module is incorporated in the middle takes the output of the bi-directional Recurrent Neural Network (Bi-RNN) module in the first stage and provides the Convolutional Neural Network (CNN) module in the last stage with the i...

Neural nets for massively parallel optimization

Science.gov (United States)

Dixon, Laurence C. W.; Mills, David

1992-07-01

To apply massively parallel processing systems to the solution of large scale optimization problems it is desirable to be able to evaluate any function f(z), z (epsilon) Rn in a parallel manner. The theorem of Cybenko, Hecht Nielsen, Hornik, Stinchcombe and White, and Funahasi shows that this can be achieved by a neural network with one hidden layer. In this paper we address the problem of the number of nodes required in the layer to achieve a given accuracy in the function and gradient values at all points within a given n dimensional interval. The type of activation function needed to obtain nonsingular Hessian matrices is described and a strategy for obtaining accurate minimal networks presented.

A hybrid model using decision tree and neural network for credit scoring problem

Directory of Open Access Journals (Sweden)

Amir Arzy Soltan

2012-08-01

Full Text Available Nowadays credit scoring is an important issue for financial and monetary organizations that has substantial impact on reduction of customer attraction risks. Identification of high risk customer can reduce finished cost. An accurate classification of customer and low type 1 and type 2 errors have been investigated in many studies. The primary objective of this paper is to develop a new method, which chooses the best neural network architecture based on one column hidden layer MLP, multiple columns hidden layers MLP, RBFN and decision trees and ensembling them with voting methods. The proposed method of this paper is run on an Australian credit data and a private bank in Iran called Export Development Bank of Iran and the results are used for making solution in low customer attraction risks.

A training rule which guarantees finite-region stability for a class of closed-loop neural-network control systems.

Science.gov (United States)

Kuntanapreeda, S; Fullmer, R R

1996-01-01

A training method for a class of neural network controllers is presented which guarantees closed-loop system stability. The controllers are assumed to be nonlinear, feedforward, sampled-data, full-state regulators implemented as single hidden-layer neural networks. The controlled systems must be locally hermitian and observable. Stability of the closed-loop system is demonstrated by determining a Lyapunov function, which can be used to identify a finite stability region about the regulator point.

A new backpropagation learning algorithm for layered neural networks with nondifferentiable units.

Science.gov (United States)

Oohori, Takahumi; Naganuma, Hidenori; Watanabe, Kazuhisa

2007-05-01

We propose a digital version of the backpropagation algorithm (DBP) for three-layered neural networks with nondifferentiable binary units. This approach feeds teacher signals to both the middle and output layers, whereas with a simple perceptron, they are given only to the output layer. The additional teacher signals enable the DBP to update the coupling weights not only between the middle and output layers but also between the input and middle layers. A neural network based on DBP learning is fast and easy to implement in hardware. Simulation results for several linearly nonseparable problems such as XOR demonstrate that the DBP performs favorably when compared to the conventional approaches. Furthermore, in large-scale networks, simulation results indicate that the DBP provides high performance.

Prediction of Industrial Electric Energy Consumption in Anhui Province Based on GA-BP Neural Network

Science.gov (United States)

Zhang, Jiajing; Yin, Guodong; Ni, Youcong; Chen, Jinlan

2018-01-01

In order to improve the prediction accuracy of industrial electrical energy consumption, a prediction model of industrial electrical energy consumption was proposed based on genetic algorithm and neural network. The model use genetic algorithm to optimize the weights and thresholds of BP neural network, and the model is used to predict the energy consumption of industrial power in Anhui Province, to improve the prediction accuracy of industrial electric energy consumption in Anhui province. By comparing experiment of GA-BP prediction model and BP neural network model, the GA-BP model is more accurate with smaller number of neurons in the hidden layer.

A learning rule for very simple universal approximators consisting of a single layer of perceptrons.

Science.gov (United States)

Auer, Peter; Burgsteiner, Harald; Maass, Wolfgang

2008-06-01

One may argue that the simplest type of neural networks beyond a single perceptron is an array of several perceptrons in parallel. In spite of their simplicity, such circuits can compute any Boolean function if one views the majority of the binary perceptron outputs as the binary output of the parallel perceptron, and they are universal approximators for arbitrary continuous functions with values in [0,1] if one views the fraction of perceptrons that output 1 as the analog output of the parallel perceptron. Note that in contrast to the familiar model of a "multi-layer perceptron" the parallel perceptron that we consider here has just binary values as outputs of gates on the hidden layer. For a long time one has thought that there exists no competitive learning algorithm for these extremely simple neural networks, which also came to be known as committee machines. It is commonly assumed that one has to replace the hard threshold gates on the hidden layer by sigmoidal gates (or RBF-gates) and that one has to tune the weights on at least two successive layers in order to achieve satisfactory learning results for any class of neural networks that yield universal approximators. We show that this assumption is not true, by exhibiting a simple learning algorithm for parallel perceptrons - the parallel delta rule (p-delta rule). In contrast to backprop for multi-layer perceptrons, the p-delta rule only has to tune a single layer of weights, and it does not require the computation and communication of analog values with high precision. Reduced communication also distinguishes our new learning rule from other learning rules for parallel perceptrons such as MADALINE. Obviously these features make the p-delta rule attractive as a biologically more realistic alternative to backprop in biological neural circuits, but also for implementations in special purpose hardware. We show that the p-delta rule also implements gradient descent-with regard to a suitable error measure

Kernel Function Tuning for Single-Layer Neural Networks

Czech Academy of Sciences Publication Activity Database

Vidnerová, Petra; Neruda, Roman

-, accepted 28.11. 2017 (2018) ISSN 2278-0149 R&D Projects: GA ČR GA15-18108S Institutional support: RVO:67985807 Keywords : single-layer neural networks * kernel methods * kernel function * optimisation Subject RIV: IN - Informatics, Computer Science http://www.ijmerr.com/

Antibacterial, anti-inflammatory and neuroprotective layer-by-layer coatings for neural implants

Science.gov (United States)

Zhang, Zhiling; Nong, Jia; Zhong, Yinghui

2015-08-01

Objective. Infection, inflammation, and neuronal loss are common issues that seriously affect the functionality and longevity of chronically implanted neural prostheses. Minocycline hydrochloride (MH) is a broad-spectrum antibiotic and effective anti-inflammatory drug that also exhibits potent neuroprotective activities. In this study, we investigated the development of biocompatible thin film coatings capable of sustained release of MH for improving the long term performance of implanted neural electrodes. Approach. We developed a novel magnesium binding-mediated drug delivery mechanism for controlled and sustained release of MH from an ultrathin hydrophilic layer-by-layer (LbL) coating and characterized the parameters that control MH loading and release. The anti-biofilm, anti-inflammatory and neuroprotective potencies of the LbL coating and released MH were also examined. Main results. Sustained release of physiologically relevant amount of MH for 46 days was achieved from the Mg2+-based LbL coating at a thickness of 1.25 μm. In addition, MH release from the LbL coating is pH-sensitive. The coating and released MH demonstrated strong anti-biofilm, anti-inflammatory, and neuroprotective potencies. Significance. This study reports, for the first time, the development of a bioactive coating that can target infection, inflammation, and neuroprotection simultaneously, which may facilitate the translation of neural interfaces to clinical applications.

Foreground removal from WMAP 5 yr temperature maps using an MLP neural network

Science.gov (United States)

Nørgaard-Nielsen, H. U.

2010-09-01

Aims: One of the main obstacles for extracting the cosmic microwave background (CMB) signal from observations in the mm/sub-mm range is the foreground contamination by emission from Galactic component: mainly synchrotron, free-free, and thermal dust emission. The statistical nature of the intrinsic CMB signal makes it essential to minimize the systematic errors in the CMB temperature determinations. Methods: The feasibility of using simple neural networks to extract the CMB signal from detailed simulated data has already been demonstrated. Here, simple neural networks are applied to the WMAP 5 yr temperature data without using any auxiliary data. Results: A simple multilayer perceptron neural network with two hidden layers provides temperature estimates over more than 75 per cent of the sky with random errors significantly below those previously extracted from these data. Also, the systematic errors, i.e. errors correlated with the Galactic foregrounds, are very small. Conclusions: With these results the neural network method is well prepared for dealing with the high - quality CMB data from the ESA Planck Surveyor satellite. unknown author type, collab

Stator current harmonics evolution by neural network method based on CFE/SS algorithm for ACEC generator of Rey Power Plant

International Nuclear Information System (INIS)

Soleymani, S.; Ranjbar, A.M.; Mirabedini, H.

2001-01-01

One method for on-line fault diagnosis in synchronous generator is stator current harmonics analysis. Then artificial neural network is considered in this paper in order to evaluate stator current harmonics in different loads. Training set of artificial neural network is made ready by generator modeling, finite element method and state space model. Many points from generator capability curve are used in order to complete this set. Artificial neural network which is used in this paper is a percept ron network with a single hidden layer, Eight hidden neurons and back propagation algorithm. Results are indicated that the trained artificial neural network can identify stator current harmonics for arbitrary load from the capability curve. The error is less than 10% in comparison with values obtained directly from the CFE-SS algorithm. The rating parameters of modeled generator are 43950 (kV A), 11(KV), 3000 (rpm), 50 (H Z), (P F=0.8)

A one-layer recurrent neural network for constrained nonsmooth optimization.

Science.gov (United States)

Liu, Qingshan; Wang, Jun

2011-10-01

This paper presents a novel one-layer recurrent neural network modeled by means of a differential inclusion for solving nonsmooth optimization problems, in which the number of neurons in the proposed neural network is the same as the number of decision variables of optimization problems. Compared with existing neural networks for nonsmooth optimization problems, the global convexity condition on the objective functions and constraints is relaxed, which allows the objective functions and constraints to be nonconvex. It is proven that the state variables of the proposed neural network are convergent to optimal solutions if a single design parameter in the model is larger than a derived lower bound. Numerical examples with simulation results substantiate the effectiveness and illustrate the characteristics of the proposed neural network.

Modeling and prediction of Turkey's electricity consumption using Artificial Neural Networks

International Nuclear Information System (INIS)

Kavaklioglu, Kadir; Ozturk, Harun Kemal; Canyurt, Olcay Ersel; Ceylan, Halim

2009-01-01

Artificial Neural Networks are proposed to model and predict electricity consumption of Turkey. Multi layer perceptron with backpropagation training algorithm is used as the neural network topology. Tangent-sigmoid and pure-linear transfer functions are selected in the hidden and output layer processing elements, respectively. These input-output network models are a result of relationships that exist among electricity consumption and several other socioeconomic variables. Electricity consumption is modeled as a function of economic indicators such as population, gross national product, imports and exports. It is also modeled using export-import ratio and time input only. Performance comparison among different models is made based on absolute and percentage mean square error. Electricity consumption of Turkey is predicted until 2027 using data from 1975 to 2006 along with other economic indicators. The results show that electricity consumption can be modeled using Artificial Neural Networks, and the models can be used to predict future electricity consumption. (author)

Growth kinetics of borided layers: Artificial neural network and least square approaches

Science.gov (United States)

Campos, I.; Islas, M.; Ramírez, G.; VillaVelázquez, C.; Mota, C.

2007-05-01

The present study evaluates the growth kinetics of the boride layer Fe 2B in AISI 1045 steel, by means of neural networks and the least square techniques. The Fe 2B phase was formed at the material surface using the paste boriding process. The surface boron potential was modified considering different boron paste thicknesses, with exposure times of 2, 4 and 6 h, and treatment temperatures of 1193, 1223 and 1273 K. The neural network and the least square models were set by the layer thickness of Fe 2B phase, and assuming that the growth of the boride layer follows a parabolic law. The reliability of the techniques used is compared with a set of experiments at a temperature of 1223 K with 5 h of treatment time and boron potentials of 2, 3, 4 and 5 mm. The results of the Fe 2B layer thicknesses show a mean error of 5.31% for the neural network and 3.42% for the least square method.

Using neural networks to describe tracer correlations

Directory of Open Access Journals (Sweden)

D. J. Lary

2004-01-01

Full Text Available Neural networks are ideally suited to describe the spatial and temporal dependence of tracer-tracer correlations. The neural network performs well even in regions where the correlations are less compact and normally a family of correlation curves would be required. For example, the CH4-N2O correlation can be well described using a neural network trained with the latitude, pressure, time of year, and methane volume mixing ratio (v.m.r.. In this study a neural network using Quickprop learning and one hidden layer with eight nodes was able to reproduce the CH4-N2O correlation with a correlation coefficient between simulated and training values of 0.9995. Such an accurate representation of tracer-tracer correlations allows more use to be made of long-term datasets to constrain chemical models. Such as the dataset from the Halogen Occultation Experiment (HALOE which has continuously observed CH4  (but not N2O from 1991 till the present. The neural network Fortran code used is available for download.

On the complexity of neural network classifiers: a comparison between shallow and deep architectures.

Science.gov (United States)

Bianchini, Monica; Scarselli, Franco

2014-08-01

Recently, researchers in the artificial neural network field have focused their attention on connectionist models composed by several hidden layers. In fact, experimental results and heuristic considerations suggest that deep architectures are more suitable than shallow ones for modern applications, facing very complex problems, e.g., vision and human language understanding. However, the actual theoretical results supporting such a claim are still few and incomplete. In this paper, we propose a new approach to study how the depth of feedforward neural networks impacts on their ability in implementing high complexity functions. First, a new measure based on topological concepts is introduced, aimed at evaluating the complexity of the function implemented by a neural network, used for classification purposes. Then, deep and shallow neural architectures with common sigmoidal activation functions are compared, by deriving upper and lower bounds on their complexity, and studying how the complexity depends on the number of hidden units and the used activation function. The obtained results seem to support the idea that deep networks actually implements functions of higher complexity, so that they are able, with the same number of resources, to address more difficult problems.

Neuroevolution Mechanism for Hidden Markov Model

Directory of Open Access Journals (Sweden)

Nabil M. Hewahi

2011-12-01

Full Text Available Hidden Markov Model (HMM is a statistical model based on probabilities. HMM is becoming one of the major models involved in many applications such as natural language
processing, handwritten recognition, image processing, prediction systems and many more. In this research we are concerned with finding out the best HMM for a certain application domain. We propose a neuroevolution process that is based first on converting the HMM to a neural network, then generating many neural networks at random where each represents a HMM. We proceed by
applying genetic operators to obtain new set of neural networks where each represents HMMs, and updating the population. Finally select the best neural network based on a fitness function.

Gradual DropIn of Layers to Train Very Deep Neural Networks

OpenAIRE

Smith, Leslie N.; Hand, Emily M.; Doster, Timothy

2015-01-01

We introduce the concept of dynamically growing a neural network during training. In particular, an untrainable deep network starts as a trainable shallow network and newly added layers are slowly, organically added during training, thereby increasing the network's depth. This is accomplished by a new layer, which we call DropIn. The DropIn layer starts by passing the output from a previous layer (effectively skipping over the newly added layers), then increasingly including units from the ne...

Thermoelectric generator hidden in a shirt with a fabric radiator

NARCIS (Netherlands)

Leonov, V.; Vullers, R.J.M.; Hoof, C.V.

2012-01-01

Integration of thermopiles in garments has been performed in this work in different ways. It is shown that textile has a minor effect on power generation, which enables completely hidden and unobtrusive energy harvester. A one-milliwatt thermoelectric generator is then integrated between two layers

A one-layer recurrent neural network for constrained nonconvex optimization.

Science.gov (United States)

Li, Guocheng; Yan, Zheng; Wang, Jun

2015-01-01

In this paper, a one-layer recurrent neural network is proposed for solving nonconvex optimization problems subject to general inequality constraints, designed based on an exact penalty function method. It is proved herein that any neuron state of the proposed neural network is convergent to the feasible region in finite time and stays there thereafter, provided that the penalty parameter is sufficiently large. The lower bounds of the penalty parameter and convergence time are also estimated. In addition, any neural state of the proposed neural network is convergent to its equilibrium point set which satisfies the Karush-Kuhn-Tucker conditions of the optimization problem. Moreover, the equilibrium point set is equivalent to the optimal solution to the nonconvex optimization problem if the objective function and constraints satisfy given conditions. Four numerical examples are provided to illustrate the performances of the proposed neural network.

Forecast of TEXT plasma disruptions using soft X rays as input signal in a neural network

International Nuclear Information System (INIS)

Vannucci, A.; Oliveira, K.A.; Tajima, T.

1999-01-01

A feedforward neural network with two hidden layers is used to forecast major and minor disruptive instabilities in TEXT tokamak discharges. Using the experimental data of soft X ray signals as input data, the neural network is trained with one disruptive plasma discharge, and a different disruptive discharge is used for validation. After being properly trained, the networks, with the same set of weights, are used to forecast disruptions in two other plasma discharges. It is observed that the neural network is able to predict the occurrence of a disruption more than 3 ms in advance. This time interval is almost 3 times longer than the one already obtained previously when a magnetic signal from a Mirnov coil was used to feed the neural networks. Visually no indication of an upcoming disruption is seen from the experimental data this far back from the time of disruption. Finally, by observing the predictive behaviour of the network for the disruptive discharges analysed and comparing the soft X ray data with the corresponding magnetic experimental signal, it is conjectured about where inside the plasma column the disruption first started. (author)

A One-Layer Recurrent Neural Network for Constrained Complex-Variable Convex Optimization.

Science.gov (United States)

Qin, Sitian; Feng, Jiqiang; Song, Jiahui; Wen, Xingnan; Xu, Chen

2018-03-01

In this paper, based on calculus and penalty method, a one-layer recurrent neural network is proposed for solving constrained complex-variable convex optimization. It is proved that for any initial point from a given domain, the state of the proposed neural network reaches the feasible region in finite time and converges to an optimal solution of the constrained complex-variable convex optimization finally. In contrast to existing neural networks for complex-variable convex optimization, the proposed neural network has a lower model complexity and better convergence. Some numerical examples and application are presented to substantiate the effectiveness of the proposed neural network.

Neutron spectrum unfolding using neural networks

International Nuclear Information System (INIS)

Vega C, H.R.; Hernandez D, V.M.; Manzanares A, E.

2004-01-01

An artificial neural network has been designed to obtain the neutron spectra from the Bonner spheres spectrometer's count rates. The neural network was trained using a large set of neutron spectra compiled by the International Atomic Energy Agency. These include spectra from iso- topic neutron sources, reference and operational neutron spectra obtained from accelerators and nuclear reactors. The spectra were transformed from lethargy to energy distribution and were re-binned to 31 energy groups using the MCNP 4C code. Re-binned spectra and UTA4 matrix were used to calculate the expected count rates in Bonner spheres spectrometer. These count rates were used as input and correspondent spectrum was used as output during neural network training. The network has 7 input nodes, 56 neurons as hidden layer and 31 neurons in the output layer. After training the network was tested with the Bonner spheres count rates produced by twelve neutron spectra. The network allows unfolding the neutron spectrum from count rates measured with Bonner spheres. Good results are obtained when testing count rates belong to neutron spectra used during training, acceptable results are obtained for count rates obtained from actual neutron fields; however the network fails when count rates belong to monoenergetic neutron sources. (Author)

Performance of Deep and Shallow Neural Networks, the Universal Approximation Theorem, Activity Cliffs, and QSAR.

Science.gov (United States)

Winkler, David A; Le, Tu C

2017-01-01

Neural networks have generated valuable Quantitative Structure-Activity/Property Relationships (QSAR/QSPR) models for a wide variety of small molecules and materials properties. They have grown in sophistication and many of their initial problems have been overcome by modern mathematical techniques. QSAR studies have almost always used so-called "shallow" neural networks in which there is a single hidden layer between the input and output layers. Recently, a new and potentially paradigm-shifting type of neural network based on Deep Learning has appeared. Deep learning methods have generated impressive improvements in image and voice recognition, and are now being applied to QSAR and QSAR modelling. This paper describes the differences in approach between deep and shallow neural networks, compares their abilities to predict the properties of test sets for 15 large drug data sets (the kaggle set), discusses the results in terms of the Universal Approximation theorem for neural networks, and describes how DNN may ameliorate or remove troublesome "activity cliffs" in QSAR data sets. © 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

Identification of a Typical CSTR Using Optimal Focused Time Lagged Recurrent Neural Network Model with Gamma Memory Filter

Directory of Open Access Journals (Sweden)

S. N. Naikwad

2009-01-01

Full Text Available A focused time lagged recurrent neural network (FTLR NN with gamma memory filter is designed to learn the subtle complex dynamics of a typical CSTR process. Continuous stirred tank reactor exhibits complex nonlinear operations where reaction is exothermic. It is noticed from literature review that process control of CSTR using neuro-fuzzy systems was attempted by many, but optimal neural network model for identification of CSTR process is not yet available. As CSTR process includes temporal relationship in the input-output mappings, time lagged recurrent neural network is particularly used for identification purpose. The standard back propagation algorithm with momentum term has been proposed in this model. The various parameters like number of processing elements, number of hidden layers, training and testing percentage, learning rule and transfer function in hidden and output layer are investigated on the basis of performance measures like MSE, NMSE, and correlation coefficient on testing data set. Finally effects of different norms are tested along with variation in gamma memory filter. It is demonstrated that dynamic NN model has a remarkable system identification capability for the problems considered in this paper. Thus FTLR NN with gamma memory filter can be used to learn underlying highly nonlinear dynamics of the system, which is a major contribution of this paper.

Neural network analysis in pharmacogenetics of mood disorders

Directory of Open Access Journals (Sweden)

Serretti Alessandro

2004-12-01

Full Text Available Abstract Background The increasing number of available genotypes for genetic studies in humans requires more advanced techniques of analysis. We previously reported significant univariate associations between gene polymorphisms and antidepressant response in mood disorders. However the combined analysis of multiple gene polymorphisms and clinical variables requires the use of non linear methods. Methods In the present study we tested a neural network strategy for a combined analysis of two gene polymorphisms. A Multi Layer Perceptron model showed the best performance and was therefore selected over the other networks. One hundred and twenty one depressed inpatients treated with fluvoxamine in the context of previously reported pharmacogenetic studies were included. The polymorphism in the transcriptional control region upstream of the 5HTT coding sequence (SERTPR and in the Tryptophan Hydroxylase (TPH gene were analysed simultaneously. Results A multi layer perceptron network composed by 1 hidden layer with 7 nodes was chosen. 77.5 % of responders and 51.2% of non responders were correctly classified (ROC area = 0.731 – empirical p value = 0.0082. Finally, we performed a comparison with traditional techniques. A discriminant function analysis correctly classified 34.1 % of responders and 68.1 % of non responders (F = 8.16 p = 0.0005. Conclusions Overall, our findings suggest that neural networks may be a valid technique for the analysis of gene polymorphisms in pharmacogenetic studies. The complex interactions modelled through NN may be eventually applied at the clinical level for the individualized therapy.

Fault Diagnosis of Hydraulic Servo Valve Based on Genetic Optimization RBF-BP Neural Network

Directory of Open Access Journals (Sweden)

Li-Ping FAN

2014-04-01

Full Text Available Electro-hydraulic servo valves are core components of the hydraulic servo system of rolling mills. It is necessary to adopt an effective fault diagnosis method to keep the hydraulic servo valve in a good work state. In this paper, RBF and BP neural network are integrated effectively to build a double hidden layers RBF-BP neural network for fault diagnosis. In the process of training the neural network, genetic algorithm (GA is used to initialize and optimize the connection weights and thresholds of the network. Several typical fault states are detected by the constructed GA-optimized fault diagnosis scheme. Simulation results shown that the proposed fault diagnosis scheme can give satisfactory effect.

Optimum operating conditions for a water purification process integrated to a heat transformer with energy recycling using neural network inverse

Energy Technology Data Exchange (ETDEWEB)

Hernandez, J.A.; Siqueiros, J.; Juarez-Romero, D. [Centro de Investigacion en Ingenieria y Ciencias Aplicadas, Universidad Autonoma del Estado de Morelos (UAEM), Av. Universidad No. 1001, Col. Chamilpa, Cuernavaca, Morelos C.P. 62209 (Mexico); Bassam, A. [Posgrado en Ingenieria y Ciencias Aplicadas, Universidad Autonoma del Estado de Morelos (UAEM), Av. Universidad No. 1001, Col. Chamilpa, Cuernavaca, Morelos C.P. 62209 (Mexico)

2009-04-15

Artificial neural network inverse (ANNi) is applied to calculate the optimal operating conditions on the coefficient of performance (COP) for a water purification process integrated to an absorption heat transformer with energy recycling. An artificial neural network (ANN) model is developed to predict the COP which was increased with energy recycling. This ANN model takes into account the input and output temperatures for each one of the four components (absorber, generator, evaporator, and condenser), as well as two pressures and LiBr + H{sub 2}O concentrations. For the network, a feedforward with one hidden layer, a Levenberg-Marquardt learning algorithm, a hyperbolic tangent sigmoid transfer function and a linear transfer function were used. The best fitting training data set was obtained with three neurons in the hidden layer. On the validation data set, simulations and experimental data test were in good agreement (R > 0.99). This ANN model can be used to predict the COP when the input variables (operating conditions) are well known. However, to control the COP in the system, we developed a strategy to estimate the optimal input variables when a COP is required from ANNi. An optimization method (the Nelder-Mead simplex method) is used to fit the unknown input variable resulted from the ANNi. This methodology can be applied to control on-line the performance of the system. (author)

Computer-assisted detection of colonic polyps with CT colonography using neural networks and binary classification trees

International Nuclear Information System (INIS)

Jerebko, Anna K.; Summers, Ronald M.; Malley, James D.; Franaszek, Marek; Johnson, C. Daniel

2003-01-01

Detection of colonic polyps in CT colonography is problematic due to complexities of polyp shape and the surface of the normal colon. Published results indicate the feasibility of computer-aided detection of polyps but better classifiers are needed to improve specificity. In this paper we compare the classification results of two approaches: neural networks and recursive binary trees. As our starting point we collect surface geometry information from three-dimensional reconstruction of the colon, followed by a filter based on selected variables such as region density, Gaussian and average curvature and sphericity. The filter returns sites that are candidate polyps, based on earlier work using detection thresholds, to which the neural nets or the binary trees are applied. A data set of 39 polyps from 3 to 25 mm in size was used in our investigation. For both neural net and binary trees we use tenfold cross-validation to better estimate the true error rates. The backpropagation neural net with one hidden layer trained with Levenberg-Marquardt algorithm achieved the best results: sensitivity 90% and specificity 95% with 16 false positives per study

Spatial Disaggregation of Areal Rainfall Using Two Different Artificial Neural Networks Models

Directory of Open Access Journals (Sweden)

Sungwon Kim

2015-06-01

Full Text Available The objective of this study is to develop artificial neural network (ANN models, including multilayer perceptron (MLP and Kohonen self-organizing feature map (KSOFM, for spatial disaggregation of areal rainfall in the Wi-stream catchment, an International Hydrological Program (IHP representative catchment, in South Korea. A three-layer MLP model, using three training algorithms, was used to estimate areal rainfall. The Levenberg–Marquardt training algorithm was found to be more sensitive to the number of hidden nodes than were the conjugate gradient and quickprop training algorithms using the MLP model. Results showed that the networks structures of 11-5-1 (conjugate gradient and quickprop and 11-3-1 (Levenberg-Marquardt were the best for estimating areal rainfall using the MLP model. The networks structures of 1-5-11 (conjugate gradient and quickprop and 1-3-11 (Levenberg–Marquardt, which are the inverse networks for estimating areal rainfall using the best MLP model, were identified for spatial disaggregation of areal rainfall using the MLP model. The KSOFM model was compared with the MLP model for spatial disaggregation of areal rainfall. The MLP and KSOFM models could disaggregate areal rainfall into individual point rainfall with spatial concepts.

Method Accelerates Training Of Some Neural Networks

Science.gov (United States)

Shelton, Robert O.

1992-01-01

Three-layer networks trained faster provided two conditions are satisfied: numbers of neurons in layers are such that majority of work done in synaptic connections between input and hidden layers, and number of neurons in input layer at least as great as number of training pairs of input and output vectors. Based on modified version of back-propagation method.

Entropy Learning in Neural Network

Directory of Open Access Journals (Sweden)

Geok See Ng

2017-12-01

Full Text Available In this paper, entropy term is used in the learning phase of a neural network.  As learning progresses, more hidden nodes get into saturation.  The early creation of such hidden nodes may impair generalisation.  Hence entropy approach is proposed to dampen the early creation of such nodes.  The entropy learning also helps to increase the importance of relevant nodes while dampening the less important nodes.  At the end of learning, the less important nodes can then be eliminated to reduce the memory requirements of the neural network.

A neural network model of the relativistic electron flux at geosynchronous orbit

International Nuclear Information System (INIS)

Koons, H.C.; Gorney, D.J.

1991-01-01

A neural network has been developed to model the temporal variations of relativistic (>3 MeV) electrons at geosynchronous orbit based on model inputs consisting of 10 consecutive days of the daily sum of the planetary magnetic index ΣKp. The neural network consists of three layers of neurons, containing 10 neurons in the input layer, 6 neurons in a hidden layer, and 1 output neuron. The output is a prediction of the daily-averaged electron flux for the tenth day. The neural network was trained using 62 days of data from July 1, 1984, through August 31, 1984, from the SEE spectrometer on the geosynchronous spacecraft 1982-019. The performance of the model was measured by comparing model outputs with measured fluxes over a 6-year period from April 19, 1982, to June 4, 1988. For the entire data set the rms logarithmic error of the neural network is 0.76, and the average logarithmic error is 0.58. The neural network is essentially zero biased, and for accumulation intervals of 3 days or longer the average logarithmic error is less than 0.1. The neural network provides results that are significantly more accurate than those from linear prediction filters. The model has been used to simulate conditions which are rarely observed in nature, such as long periods of quiet (ΣKp = 0) and ideal impulses. It has also been used to make reasonably accurate day-ahead forecasts of the relativistic electron flux at geosynchronous orbit

Application of neural networks to prediction of phase transport characteristics in high-pressure two-phase turbulent bubbly flows

International Nuclear Information System (INIS)

Yang, A.-S.; Kuo, T.-C.; Ling, P.-H.

2003-01-01

The phase transport phenomenon of the high-pressure two-phase turbulent bubbly flow involves complicated interfacial interactions of the mass, momentum, and energy transfer processes between phases, revealing that an enormous effort is required in characterizing the liquid-gas flow behavior. Nonetheless, the instantaneous information of bubbly flow properties is often desired for many industrial applications. This investigation aims to demonstrate the successful use of neural networks in the real-time determination of two-phase flow properties at elevated pressures. Three back-propagation neural networks, trained with the simulation results of a comprehensive theoretical model, are established to predict the transport characteristics (specifically the distributions of void-fraction and axial liquid-gas velocities) of upward turbulent bubbly pipe flows at pressures covering 3.5-7.0 MPa. Comparisons of the predictions with the test target vectors indicate that the averaged root-mean-squared (RMS) error for each one of three back-propagation neural networks is within 4.59%. In addition, this study appraises the effects of different network parameters, including the number of hidden nodes, the type of transfer function, the number of training pairs, the learning rate-increasing ratio, the learning rate-decreasing ratio, and the momentum value, on the training quality of neural networks.

Application of backpropagation neural networks to evaluate residual properties of thermally damaged concrete

International Nuclear Information System (INIS)

Vasconcelos, W.L.; Shigaki, Y.; Tolentino, E.

2009-01-01

In this work it was analyzed the residual performance of Portland cement concretes, when cold after heat-treated up to 600 deg C. Granite-gneiss was used in the three concrete mix proportions as the coarse aggregate, and river sand with finesses modulus of 2.7 as the fine aggregate. Ultrasonic pulse tests were performed on all the specimens and ultrasonic dynamic modulus were obtained. An artificial neural network of the backpropagation type was trained to evaluate and apply models in predicting residual properties of Portland cement concretes. The input layer for both models consists of an external layer input vector of the temperature. The hidden layer has two processing units with hyperbolic tangent sigmoid transfer functions (tansig for short), and the output layer contains one processing unit that represents the network's output (ultrasonic pulse velocity or modulus of elasticity) for each input vector. The training phase of the network converged for reasonable results after 5.000 epochs approximately, resulting in mean squared errors less than 0.02 for the normalized data. The neural network developed for modeling residual properties of Portland cement concretes was shown to be efficient in both the training phase and the test. From the results reasonable predictions could be made for the ultrasonic pulse velocity or dynamic modulus of elasticity by using temperature. (author)

Development of a neural network technique for KSTAR Thomson scattering diagnostics

Energy Technology Data Exchange (ETDEWEB)

Lee, Seung Hun, E-mail: leesh81@nfri.re.kr; Lee, J. H. [National Fusion Research Institute, 169-148 Gwahak-ro, Yuseong-gu, Daejeon 34133 (Korea, Republic of); Yamada, I. [National Institute Fusion Science, Toki, Gifu 509-5292 (Japan); Park, Jae Sun [Department of Physics, Korea Advanced Institute of Science and Technology, 291 Daehak-ro, Yuseong-gu, Daejeon 34141 (Korea, Republic of)

2016-11-15

Neural networks provide powerful approaches of dealing with nonlinear data and have been successfully applied to fusion plasma diagnostics and control systems. Controlling tokamak plasmas in real time is essential to measure the plasma parameters in situ. However, the χ{sup 2} method traditionally used in Thomson scattering diagnostics hampers real-time measurement due to the complexity of the calculations involved. In this study, we applied a neural network approach to Thomson scattering diagnostics in order to calculate the electron temperature, comparing the results to those obtained with the χ{sup 2} method. The best results were obtained for 10{sup 3} training cycles and eight nodes in the hidden layer. Our neural network approach shows good agreement with the χ{sup 2} method and performs the calculation twenty times faster.

Neural Networks

International Nuclear Information System (INIS)

Smith, Patrick I.

2003-01-01

information [2]. Each one of these cells acts as a simple processor. When individual cells interact with one another, the complex abilities of the brain are made possible. In neural networks, the input or data are processed by a propagation function that adds up the values of all the incoming data. The ending value is then compared with a threshold or specific value. The resulting value must exceed the activation function value in order to become output. The activation function is a mathematical function that a neuron uses to produce an output referring to its input value. [8] Figure 1 depicts this process. Neural networks usually have three components an input, a hidden, and an output. These layers create the end result of the neural network. A real world example is a child associating the word dog with a picture. The child says dog and simultaneously looks a picture of a dog. The input is the spoken word ''dog'', the hidden is the brain processing, and the output will be the category of the word dog based on the picture. This illustration describes how a neural network functions

Optimization of neural network architecture for classification of radar jamming FM signals

Science.gov (United States)

Soto, Alberto; Mendoza, Ariadna; Flores, Benjamin C.

2017-05-01

The purpose of this study is to investigate several artificial Neural Network (NN) architectures in order to design a cognitive radar system capable of optimally distinguishing linear Frequency-Modulated (FM) signals from bandlimited Additive White Gaussian Noise (AWGN). The goal is to create a theoretical framework to determine an optimal NN architecture to achieve a Probability of Detection (PD) of 95% or higher and a Probability of False Alarm (PFA) of 1.5% or lower at 5 dB Signal to Noise Ratio (SNR). Literature research reveals that the frequency-domain power spectral densities characterize a signal more efficiently than its time-domain counterparts. Therefore, the input data is preprocessed by calculating the magnitude square of the Discrete Fourier Transform of the digitally sampled bandlimited AWGN and linear FM signals to populate a matrix containing N number of samples and M number of spectra. This matrix is used as input for the NN, and the spectra are divided as follows: 70% for training, 15% for validation, and 15% for testing. The study begins by experimentally deducing the optimal number of hidden neurons (1-40 neurons), then the optimal number of hidden layers (1-5 layers), and lastly, the most efficient learning algorithm. The training algorithms examined are: Resilient Backpropagation, Scaled Conjugate Gradient, Conjugate Gradient with Powell/Beale Restarts, Polak-Ribiére Conjugate Gradient, and Variable Learning Rate Backpropagation. We determine that an architecture with ten hidden neurons (or higher), one hidden layer, and a Scaled Conjugate Gradient for training algorithm encapsulates an optimal architecture for our application.

Fast, Simple and Accurate Handwritten Digit Classification by Training Shallow Neural Network Classifiers with the 'Extreme Learning Machine' Algorithm.

Directory of Open Access Journals (Sweden)

Mark D McDonnell

Full Text Available Recent advances in training deep (multi-layer architectures have inspired a renaissance in neural network use. For example, deep convolutional networks are becoming the default option for difficult tasks on large datasets, such as image and speech recognition. However, here we show that error rates below 1% on the MNIST handwritten digit benchmark can be replicated with shallow non-convolutional neural networks. This is achieved by training such networks using the 'Extreme Learning Machine' (ELM approach, which also enables a very rapid training time (∼ 10 minutes. Adding distortions, as is common practise for MNIST, reduces error rates even further. Our methods are also shown to be capable of achieving less than 5.5% error rates on the NORB image database. To achieve these results, we introduce several enhancements to the standard ELM algorithm, which individually and in combination can significantly improve performance. The main innovation is to ensure each hidden-unit operates only on a randomly sized and positioned patch of each image. This form of random 'receptive field' sampling of the input ensures the input weight matrix is sparse, with about 90% of weights equal to zero. Furthermore, combining our methods with a small number of iterations of a single-batch backpropagation method can significantly reduce the number of hidden-units required to achieve a particular performance. Our close to state-of-the-art results for MNIST and NORB suggest that the ease of use and accuracy of the ELM algorithm for designing a single-hidden-layer neural network classifier should cause it to be given greater consideration either as a standalone method for simpler problems, or as the final classification stage in deep neural networks applied to more difficult problems.

Study on the identifying of meat's visible spectrum based on BP artificial neural network

Science.gov (United States)

Li, Xiaotian; Zhang, Tieqiang; Li, Bo; Jiang, Yongheng; Liu, Binghui; Li, Zhaokai

2006-01-01

A method to identify different meat by the visible and reflected spectra of meat with BP artificial neural net (BP-ANN) was introduced in this paper. The visible and reflected spectra (from 420 to 535nm) of different meat (beef and pork) were measured with fiber sensor spectrometer. A kind of ANN with a double-hidden layer was created to identify the different meat automatically. Its right ratio reaches 92.71%.

Planning Training Loads for the 400 M Hurdles in Three-Month Mesocycles using Artificial Neural Networks.

Science.gov (United States)

Przednowek, Krzysztof; Iskra, Janusz; Wiktorowicz, Krzysztof; Krzeszowski, Tomasz; Maszczyk, Adam

2017-12-01

This paper presents a novel approach to planning training loads in hurdling using artificial neural networks. The neural models performed the task of generating loads for athletes' training for the 400 meters hurdles. All the models were calculated based on the training data of 21 Polish National Team hurdlers, aged 22.25 ± 1.96, competing between 1989 and 2012. The analysis included 144 training plans that represented different stages in the annual training cycle. The main contribution of this paper is to develop neural models for planning training loads for the entire career of a typical hurdler. In the models, 29 variables were used, where four characterized the runner and 25 described the training process. Two artificial neural networks were used: a multi-layer perceptron and a network with radial basis functions. To assess the quality of the models, the leave-one-out cross-validation method was used in which the Normalized Root Mean Squared Error was calculated. The analysis shows that the method generating the smallest error was the radial basis function network with nine neurons in the hidden layer. Most of the calculated training loads demonstrated a non-linear relationship across the entire competitive period. The resulting model can be used as a tool to assist a coach in planning training loads during a selected training period.

Planning Training Loads for The 400 M Hurdles in Three-Month Mesocycles Using Artificial Neural Networks

Directory of Open Access Journals (Sweden)

Przednowek Krzysztof

2017-12-01

Full Text Available This paper presents a novel approach to planning training loads in hurdling using artificial neural networks. The neural models performed the task of generating loads for athletes’ training for the 400 meters hurdles. All the models were calculated based on the training data of 21 Polish National Team hurdlers, aged 22.25 ± 1.96, competing between 1989 and 2012. The analysis included 144 training plans that represented different stages in the annual training cycle. The main contribution of this paper is to develop neural models for planning training loads for the entire career of a typical hurdler. In the models, 29 variables were used, where four characterized the runner and 25 described the training process. Two artificial neural networks were used: a multi-layer perceptron and a network with radial basis functions. To assess the quality of the models, the leave-one-out cross-validation method was used in which the Normalized Root Mean Squared Error was calculated. The analysis shows that the method generating the smallest error was the radial basis function network with nine neurons in the hidden layer. Most of the calculated training loads demonstrated a non-linear relationship across the entire competitive period. The resulting model can be used as a tool to assist a coach in planning training loads during a selected training period.

Genetic algorithms used to optimize an artificial neural network design used in neutron spectrometry; Algoritmos geneticos utilizados para optimizar un diseno de red neuronal artificial usado en espectrometria de neutrones

Energy Technology Data Exchange (ETDEWEB)

Arteaga A, T.; Ortiz R, J. M.; Vega C, H. R., E-mail: tarcicio70@yahoo.co.uk [Universidad Autonoma de Zacatecas, Av. Lopez Velarde 117, 98600 Zacatecas, Zac. (Mexico)

2016-10-15

Artificial neural networks (Ann) are widely used; it which consist of an input layer, one or more hidden layers and an output layer; these layers contain neurons and each has connections called weights, where the knowledge are allowed and let to Ann solve problems proposed. These Ann is used to reconstruction of the energy spectrum of neutrons from count rates and develop Bonner sphere neutron dosimetry. Currently, we have developed Ann with high performance and generalization ability. Determine your optimal architecture is usually a difficult task, an exhaustive search of all possible combinations of parameters is rarely possible further training of the neural network with random initial weights can cause two major drawbacks: it can stuck in local minima or converge very slowly. In this project it will be used Genetic Algorithms (Ga); which are based on the principle or analogy of evolution through natural selection and has shown to be very effective in optimizing complex search functions and large spaces or to find a near optimal overall efficiency. The aim is to decrease the architecture in number of hidden neurons and therefore the total number of connections is reducing. The benefits obtained by optimizing the network are that the number of connections would be considerably smaller and thus the computational complexity, hardware integration, resources will be lower such that will allow to be even more viable implemented. To use the Ga three problems must be solve: 1) coding the problem into chromosomes. 2) Construct a fitness function. 3) Proper selection of genetic operators; crossover, selection, mutation. As a result, the scientific knowledge obtained can to be applied to similar problems having a reference parameters used and their impact on the optimization would to be generated. It concluded that the input layer and output are subject to the problem; the Ga propose the optimal number of neurons in the hidden layer without losing the quality of the

Supervised artificial neural network-based method for conversion of solar radiation data (case study: Algeria)

Science.gov (United States)

Laidi, Maamar; Hanini, Salah; Rezrazi, Ahmed; Yaiche, Mohamed Redha; El Hadj, Abdallah Abdallah; Chellali, Farouk

2017-04-01

In this study, a backpropagation artificial neural network (BP-ANN) model is used as an alternative approach to predict solar radiation on tilted surfaces (SRT) using a number of variables involved in physical process. These variables are namely the latitude of the site, mean temperature and relative humidity, Linke turbidity factor and Angstrom coefficient, extraterrestrial solar radiation, solar radiation data measured on horizontal surfaces (SRH), and solar zenith angle. Experimental solar radiation data from 13 stations spread all over Algeria around the year (2004) were used for training/validation and testing the artificial neural networks (ANNs), and one station was used to make the interpolation of the designed ANN. The ANN model was trained, validated, and tested using 60, 20, and 20 % of all data, respectively. The configuration 8-35-1 (8 inputs, 35 hidden, and 1 output neurons) presented an excellent agreement between the prediction and the experimental data during the test stage with determination coefficient of 0.99 and root meat squared error of 5.75 Wh/m2, considering a three-layer feedforward backpropagation neural network with Levenberg-Marquardt training algorithm, a hyperbolic tangent sigmoid and linear transfer function at the hidden and the output layer, respectively. This novel model could be used by researchers or scientists to design high-efficiency solar devices that are usually tilted at an optimum angle to increase the solar incident on the surface.

Evolutionary neural network modeling for software cumulative failure time prediction

International Nuclear Information System (INIS)

Tian Liang; Noore, Afzel

2005-01-01

An evolutionary neural network modeling approach for software cumulative failure time prediction based on multiple-delayed-input single-output architecture is proposed. Genetic algorithm is used to globally optimize the number of the delayed input neurons and the number of neurons in the hidden layer of the neural network architecture. Modification of Levenberg-Marquardt algorithm with Bayesian regularization is used to improve the ability to predict software cumulative failure time. The performance of our proposed approach has been compared using real-time control and flight dynamic application data sets. Numerical results show that both the goodness-of-fit and the next-step-predictability of our proposed approach have greater accuracy in predicting software cumulative failure time compared to existing approaches

Prediction of strain values in reinforcements and concrete of a RC frame using neural networks

Science.gov (United States)

Vafaei, Mohammadreza; Alih, Sophia C.; Shad, Hossein; Falah, Ali; Halim, Nur Hajarul Falahi Abdul

2018-03-01

The level of strain in structural elements is an important indicator for the presence of damage and its intensity. Considering this fact, often structural health monitoring systems employ strain gauges to measure strains in critical elements. However, because of their sensitivity to the magnetic fields, inadequate long-term durability especially in harsh environments, difficulties in installation on existing structures, and maintenance cost, installation of strain gauges is not always possible for all structural components. Therefore, a reliable method that can accurately estimate strain values in critical structural elements is necessary for damage identification. In this study, a full-scale test was conducted on a planar RC frame to investigate the capability of neural networks for predicting the strain values. Two neural networks each of which having a single hidden layer was trained to relate the measured rotations and vertical displacements of the frame to the strain values measured at different locations of the frame. Results of trained neural networks indicated that they accurately estimated the strain values both in reinforcements and concrete. In addition, the trained neural networks were capable of predicting strains for the unseen input data set.

Crop Classification by Forward Neural Network with Adaptive Chaotic Particle Swarm Optimization

Directory of Open Access Journals (Sweden)

Yudong Zhang

2011-05-01

Full Text Available This paper proposes a hybrid crop classifier for polarimetric synthetic aperture radar (SAR images. The feature sets consisted of span image, the H/A/α decomposition, and the gray-level co-occurrence matrix (GLCM based texture features. Then, the features were reduced by principle component analysis (PCA. Finally, a two-hidden-layer forward neural network (NN was constructed and trained by adaptive chaotic particle swarm optimization (ACPSO. K-fold cross validation was employed to enhance generation. The experimental results on Flevoland sites demonstrate the superiority of ACPSO to back-propagation (BP, adaptive BP (ABP, momentum BP (MBP, Particle Swarm Optimization (PSO, and Resilient back-propagation (RPROP methods. Moreover, the computation time for each pixel is only 1.08 × 10−7 s.

Hermite Functional Link Neural Network for Solving the Van der Pol-Duffing Oscillator Equation.

Science.gov (United States)

Mall, Susmita; Chakraverty, S

2016-08-01

Hermite polynomial-based functional link artificial neural network (FLANN) is proposed here to solve the Van der Pol-Duffing oscillator equation. A single-layer hermite neural network (HeNN) model is used, where a hidden layer is replaced by expansion block of input pattern using Hermite orthogonal polynomials. A feedforward neural network model with the unsupervised error backpropagation principle is used for modifying the network parameters and minimizing the computed error function. The Van der Pol-Duffing and Duffing oscillator equations may not be solved exactly. Here, approximate solutions of these types of equations have been obtained by applying the HeNN model for the first time. Three mathematical example problems and two real-life application problems of Van der Pol-Duffing oscillator equation, extracting the features of early mechanical failure signal and weak signal detection problems, are solved using the proposed HeNN method. HeNN approximate solutions have been compared with results obtained by the well known Runge-Kutta method. Computed results are depicted in term of graphs. After training the HeNN model, we may use it as a black box to get numerical results at any arbitrary point in the domain. Thus, the proposed HeNN method is efficient. The results reveal that this method is reliable and can be applied to other nonlinear problems too.

Ensemble learning in fixed expansion layer networks for mitigating catastrophic forgetting.

Science.gov (United States)

Coop, Robert; Mishtal, Aaron; Arel, Itamar

2013-10-01

Catastrophic forgetting is a well-studied attribute of most parameterized supervised learning systems. A variation of this phenomenon, in the context of feedforward neural networks, arises when nonstationary inputs lead to loss of previously learned mappings. The majority of the schemes proposed in the literature for mitigating catastrophic forgetting were not data driven and did not scale well. We introduce the fixed expansion layer (FEL) feedforward neural network, which embeds a sparsely encoding hidden layer to help mitigate forgetting of prior learned representations. In addition, we investigate a novel framework for training ensembles of FEL networks, based on exploiting an information-theoretic measure of diversity between FEL learners, to further control undesired plasticity. The proposed methodology is demonstrated on a basic classification task, clearly emphasizing its advantages over existing techniques. The architecture proposed can be enhanced to address a range of computational intelligence tasks, such as regression problems and system control.

Artificial Neural Network applied to lightning flashes

Science.gov (United States)

Gin, R. B.; Guedes, D.; Bianchi, R.

2013-05-01

The development of video cameras enabled cientists to study lightning discharges comportment with more precision. The main goal of this project is to create a system able to detect images of lightning discharges stored in videos and classify them using an Artificial Neural Network (ANN)using C Language and OpenCV libraries. The developed system, can be split in two different modules: detection module and classification module. The detection module uses OpenCV`s computer vision libraries and image processing techniques to detect if there are significant differences between frames in a sequence, indicating that something, still not classified, occurred. Whenever there is a significant difference between two consecutive frames, two main algorithms are used to analyze the frame image: brightness and shape algorithms. These algorithms detect both shape and brightness of the event, removing irrelevant events like birds, as well as detecting the relevant events exact position, allowing the system to track it over time. The classification module uses a neural network to classify the relevant events as horizontal or vertical lightning, save the event`s images and calculates his number of discharges. The Neural Network was implemented using the backpropagation algorithm, and was trained with 42 training images , containing 57 lightning events (one image can have more than one lightning). TheANN was tested with one to five hidden layers, with up to 50 neurons each. The best configuration achieved a success rate of 95%, with one layer containing 20 neurons (33 test images with 42 events were used in this phase). This configuration was implemented in the developed system to analyze 20 video files, containing 63 lightning discharges previously manually detected. Results showed that all the lightning discharges were detected, many irrelevant events were unconsidered, and the event's number of discharges was correctly computed. The neural network used in this project achieved a

The Multi-Layered Perceptrons Neural Networks for the Prediction of Daily Solar Radiation

OpenAIRE

Radouane Iqdour; Abdelouhab Zeroual

2007-01-01

The Multi-Layered Perceptron (MLP) Neural networks have been very successful in a number of signal processing applications. In this work we have studied the possibilities and the met difficulties in the application of the MLP neural networks for the prediction of daily solar radiation data. We have used the Polack-Ribière algorithm for training the neural networks. A comparison, in term of the statistical indicators, with a linear model most used in literature, is also perfo...

SYNAPTIC DEPRESSION IN DEEP NEURAL NETWORKS FOR SPEECH PROCESSING.

Science.gov (United States)

Zhang, Wenhao; Li, Hanyu; Yang, Minda; Mesgarani, Nima

2016-03-01

A characteristic property of biological neurons is their ability to dynamically change the synaptic efficacy in response to variable input conditions. This mechanism, known as synaptic depression, significantly contributes to the formation of normalized representation of speech features. Synaptic depression also contributes to the robust performance of biological systems. In this paper, we describe how synaptic depression can be modeled and incorporated into deep neural network architectures to improve their generalization ability. We observed that when synaptic depression is added to the hidden layers of a neural network, it reduces the effect of changing background activity in the node activations. In addition, we show that when synaptic depression is included in a deep neural network trained for phoneme classification, the performance of the network improves under noisy conditions not included in the training phase. Our results suggest that more complete neuron models may further reduce the gap between the biological performance and artificial computing, resulting in networks that better generalize to novel signal conditions.

Typology of nonlinear activity waves in a layered neural continuum.

Science.gov (United States)

Koch, Paul; Leisman, Gerry

2006-04-01

Neural tissue, a medium containing electro-chemical energy, can amplify small increments in cellular activity. The growing disturbance, measured as the fraction of active cells, manifests as propagating waves. In a layered geometry with a time delay in synaptic signals between the layers, the delay is instrumental in determining the amplified wavelengths. The growth of the waves is limited by the finite number of neural cells in a given region of the continuum. As wave growth saturates, the resulting activity patterns in space and time show a variety of forms, ranging from regular monochromatic waves to highly irregular mixtures of different spatial frequencies. The type of wave configuration is determined by a number of parameters, including alertness and synaptic conditioning as well as delay. For all cases studied, using numerical solution of the nonlinear Wilson-Cowan (1973) equations, there is an interval in delay in which the wave mixing occurs. As delay increases through this interval, during a series of consecutive waves propagating through a continuum region, the activity within that region changes from a single-frequency to a multiple-frequency pattern and back again. The diverse spatio-temporal patterns give a more concrete form to several metaphors advanced over the years to attempt an explanation of cognitive phenomena: Activity waves embody the "holographic memory" (Pribram, 1991); wave mixing provides a plausible cause of the competition called "neural Darwinism" (Edelman, 1988); finally the consecutive generation of growing neural waves can explain the discontinuousness of "psychological time" (Stroud, 1955).

New S-box calculation approach for Rijndael-AES based on an artificial neural network

Directory of Open Access Journals (Sweden)

Jaime David Rios Arrañaga

2017-11-01

Full Text Available The S-box is a basic important component in symmetric key encryption, used in block ciphers to confuse or hide the relationship between the plaintext and the ciphertext. In this paper a way to develop the transformation of an input of the S-box specified in AES encryption system through an artificial neural network and the multiplicative inverse in Galois Field is presented. With this implementation more security is achieved since the values of the S-box remain hidden and the inverse table serves as a distractor since it would appear to be the complete S-box. This is implemented on MATLAB and HSPICE using a network of perceptron neurons with a hidden layer and null error.

Analysis of Artificial Neural Network in Erosion Modeling: A Case Study of Serang Watershed

Science.gov (United States)

Arif, N.; Danoedoro, P.; Hartono

2017-12-01

Erosion modeling is an important measuring tool for both land users and decision makers to evaluate land cultivation and thus it is necessary to have a model to represent the actual reality. Erosion models are a complex model because of uncertainty data with different sources and processing procedures. Artificial neural networks can be relied on for complex and non-linear data processing such as erosion data. The main difficulty in artificial neural network training is the determination of the value of each network input parameters, i.e. hidden layer, momentum, learning rate, momentum, and RMS. This study tested the capability of artificial neural network application in the prediction of erosion risk with some input parameters through multiple simulations to get good classification results. The model was implemented in Serang Watershed, Kulonprogo, Yogyakarta which is one of the critical potential watersheds in Indonesia. The simulation results showed the number of iterations that gave a significant effect on the accuracy compared to other parameters. A small number of iterations can produce good accuracy if the combination of other parameters was right. In this case, one hidden layer was sufficient to produce good accuracy. The highest training accuracy achieved in this study was 99.32%, occurred in ANN 14 simulation with combination of network input parameters of 1 HL; LR 0.01; M 0.5; RMS 0.0001, and the number of iterations of 15000. The ANN training accuracy was not influenced by the number of channels, namely input dataset (erosion factors) as well as data dimensions, rather it was determined by changes in network parameters.

A lithology identification method for continental shale oil reservoir based on BP neural network

Science.gov (United States)

Han, Luo; Fuqiang, Lai; Zheng, Dong; Weixu, Xia

2018-06-01

The Dongying Depression and Jiyang Depression of the Bohai Bay Basin consist of continental sedimentary facies with a variable sedimentary environment and the shale layer system has a variety of lithologies and strong heterogeneity. It is difficult to accurately identify the lithologies with traditional lithology identification methods. The back propagation (BP) neural network was used to predict the lithology of continental shale oil reservoirs. Based on the rock slice identification, x-ray diffraction bulk rock mineral analysis, scanning electron microscope analysis, and the data of well logging and logging, the lithology was divided with carbonate, clay and felsic as end-member minerals. According to the core-electrical relationship, the frequency histogram was then used to calculate the logging response range of each lithology. The lithology-sensitive curves selected from 23 logging curves (GR, AC, CNL, DEN, etc) were chosen as the input variables. Finally, the BP neural network training model was established to predict the lithology. The lithology in the study area can be divided into four types: mudstone, lime mudstone, lime oil-mudstone, and lime argillaceous oil-shale. The logging responses of lithology were complicated and characterized by the low values of four indicators and medium values of two indicators. By comparing the number of hidden nodes and the number of training times, we found that the number of 15 hidden nodes and 1000 times of training yielded the best training results. The optimal neural network training model was established based on the above results. The lithology prediction results of BP neural network of well XX-1 showed that the accuracy rate was over 80%, indicating that the method was suitable for lithology identification of continental shale stratigraphy. The study provided the basis for the reservoir quality and oily evaluation of continental shale reservoirs and was of great significance to shale oil and gas exploration.

Adaptive Control of Nonlinear Discrete-Time Systems by Using OS-ELM Neural Networks

Directory of Open Access Journals (Sweden)

Xiao-Li Li

2014-01-01

Full Text Available As a kind of novel feedforward neural network with single hidden layer, ELM (extreme learning machine neural networks are studied for the identification and control of nonlinear dynamic systems. The property of simple structure and fast convergence of ELM can be shown clearly. In this paper, we are interested in adaptive control of nonlinear dynamic plants by using OS-ELM (online sequential extreme learning machine neural networks. Based on data scope division, the problem that training process of ELM neural network is sensitive to the initial training data is also solved. According to the output range of the controlled plant, the data corresponding to this range will be used to initialize ELM. Furthermore, due to the drawback of conventional adaptive control, when the OS-ELM neural network is used for adaptive control of the system with jumping parameters, the topological structure of the neural network can be adjusted dynamically by using multiple model switching strategy, and an MMAC (multiple model adaptive control will be used to improve the control performance. Simulation results are included to complement the theoretical results.

Use of neural network based auto-associative memory as a data compressor for pre-processing optical emission spectra in gas thermometry with the help of neural network

International Nuclear Information System (INIS)

Dolenko, S.A.; Filippov, A.V.; Pal, A.F.; Persiantsev, I.G.; Serov, A.O.

2003-01-01

Determination of temperature from optical emission spectra is an inverse problem that is often very difficult to solve, especially when substantial noise is present. One of the means that can be used to solve such a problem is a neural network trained on the results of modeling of spectra at different temperatures (Dolenko, et al., in: I.C. Parmee (Ed.), Adaptive Computing in Design and Manufacture, Springer, London, 1998, p. 345). Reducing the dimensionality of the input data prior to application of neural network can increase the accuracy and stability of temperature determination. In this study, such pre-processing is performed with another neural network working as an auto-associative memory with a narrow bottleneck in the hidden layer. The improvement in the accuracy and stability of temperature determination in presence of noise is demonstrated on model spectra similar to those recorded in a DC-discharge CVD reactor

Basic problems solving for two-dimensional discrete 3 × 4 order hidden markov model

International Nuclear Information System (INIS)

Wang, Guo-gang; Gan, Zong-liang; Tang, Gui-jin; Cui, Zi-guan; Zhu, Xiu-chang

2016-01-01

A novel model is proposed to overcome the shortages of the classical hypothesis of the two-dimensional discrete hidden Markov model. In the proposed model, the state transition probability depends on not only immediate horizontal and vertical states but also on immediate diagonal state, and the observation symbol probability depends on not only current state but also on immediate horizontal, vertical and diagonal states. This paper defines the structure of the model, and studies the three basic problems of the model, including probability calculation, path backtracking and parameters estimation. By exploiting the idea that the sequences of states on rows or columns of the model can be seen as states of a one-dimensional discrete 1 × 2 order hidden Markov model, several algorithms solving the three questions are theoretically derived. Simulation results further demonstrate the performance of the algorithms. Compared with the two-dimensional discrete hidden Markov model, there are more statistical characteristics in the structure of the proposed model, therefore the proposed model theoretically can more accurately describe some practical problems.

Comparing Two Methods of Neural Networks to Evaluate Dead Oil Viscosity

Directory of Open Access Journals (Sweden)

Meysam Dabiri-Atashbeyk

2018-01-01

Full Text Available Reservoir characterization and asset management require comprehensive information about formation fluids. In fact, it is not possible to find accurate solutions to many petroleum engineering problems without having accurate pressure-volume-temperature (PVT data. Traditionally, fluid information has been obtained by capturing samples and then by measuring the PVT properties in a laboratory. In recent years, neural network has been applied to a large number of petroleum engineering problems. In this paper, a multi-layer perception neural network and radial basis function network (both optimized by a genetic algorithm were used to evaluate the dead oil viscosity of crude oil, and it was found out that the estimated dead oil viscosity by the multi-layer perception neural network was more accurate than the one obtained by radial basis function network.

THE USE OF NEURAL NETWORK TECHNOLOGY TO MODEL SWIMMING PERFORMANCE

Directory of Open Access Journals (Sweden)

António José Silva

2007-03-01

Full Text Available The aims of the present study were: to identify the factors which are able to explain the performance in the 200 meters individual medley and 400 meters front crawl events in young swimmers, to model the performance in those events using non-linear mathematic methods through artificial neural networks (multi-layer perceptrons and to assess the neural network models precision to predict the performance. A sample of 138 young swimmers (65 males and 73 females of national level was submitted to a test battery comprising four different domains: kinanthropometric evaluation, dry land functional evaluation (strength and flexibility, swimming functional evaluation (hydrodynamics, hydrostatic and bioenergetics characteristics and swimming technique evaluation. To establish a profile of the young swimmer non-linear combinations between preponderant variables for each gender and swim performance in the 200 meters medley and 400 meters font crawl events were developed. For this purpose a feed forward neural network was used (Multilayer Perceptron with three neurons in a single hidden layer. The prognosis precision of the model (error lower than 0.8% between true and estimated performances is supported by recent evidence. Therefore, we consider that the neural network tool can be a good approach in the resolution of complex problems such as performance modeling and the talent identification in swimming and, possibly, in a wide variety of sports

Modelling and Predicting the Breaking Strength and Mass Irregularity of Cotton Rotor-Spun Yarns Containing Cotton Fiber Recovered from Ginning Process by Using Artificial Neural Network Algorithm

Directory of Open Access Journals (Sweden)

Mohsen Shanbeh

2011-01-01

Full Text Available One of the main methods to reduce the production costs is waste recycling which is the most important challenge for the future. Cotton wastes collected from ginning process have desirable properties which could be used during spinning process. The purpose of this study was to develop predictive models of breaking strength and mass irregularity (CV% of cotton waste rotor-spun yarns containing cotton waste collected from ginning process by using the artificial neural network trained with backpropagation algorithm. Artificial neural network models have been developed based on rotor diameter, rotor speed, navel type, opener roller speed, ginning waste proportion and yarn linear density as input parameters. The parameters of artificial neural network model, namely, learning, and momentum rate, number of hidden layers and number of hidden processing elements (neurons were optimized to get the best predictive models. The findings showed that the breaking strength and mass irregularity of rotor spun yarns could be predicted satisfactorily by artificial neural network. The maximum error in predicting the breaking strength and mass irregularity of testing data was 8.34% and 6.65%, respectively.

Multi-objective evolutionary optimization for constructing neural networks for virtual reality visual data mining: application to geophysical prospecting.

Science.gov (United States)

Valdés, Julio J; Barton, Alan J

2007-05-01

A method for the construction of virtual reality spaces for visual data mining using multi-objective optimization with genetic algorithms on nonlinear discriminant (NDA) neural networks is presented. Two neural network layers (the output and the last hidden) are used for the construction of simultaneous solutions for: (i) a supervised classification of data patterns and (ii) an unsupervised similarity structure preservation between the original data matrix and its image in the new space. A set of spaces are constructed from selected solutions along the Pareto front. This strategy represents a conceptual improvement over spaces computed by single-objective optimization. In addition, genetic programming (in particular gene expression programming) is used for finding analytic representations of the complex mappings generating the spaces (a composition of NDA and orthogonal principal components). The presented approach is domain independent and is illustrated via application to the geophysical prospecting of caves.

An effective convolutional neural network model for Chinese sentiment analysis

Science.gov (United States)

Zhang, Yu; Chen, Mengdong; Liu, Lianzhong; Wang, Yadong

2017-06-01

Nowadays microblog is getting more and more popular. People are increasingly accustomed to expressing their opinions on Twitter, Facebook and Sina Weibo. Sentiment analysis of microblog has received significant attention, both in academia and in industry. So far, Chinese microblog exploration still needs lots of further work. In recent years CNN has also been used to deal with NLP tasks, and already achieved good results. However, these methods ignore the effective use of a large number of existing sentimental resources. For this purpose, we propose a Lexicon-based Sentiment Convolutional Neural Networks (LSCNN) model focus on Weibo's sentiment analysis, which combines two CNNs, trained individually base on sentiment features and word embedding, at the fully connected hidden layer. The experimental results show that our model outperforms the CNN model only with word embedding features on microblog sentiment analysis task.

A Multiobjective Sparse Feature Learning Model for Deep Neural Networks.

Science.gov (United States)

Gong, Maoguo; Liu, Jia; Li, Hao; Cai, Qing; Su, Linzhi

2015-12-01

Hierarchical deep neural networks are currently popular learning models for imitating the hierarchical architecture of human brain. Single-layer feature extractors are the bricks to build deep networks. Sparse feature learning models are popular models that can learn useful representations. But most of those models need a user-defined constant to control the sparsity of representations. In this paper, we propose a multiobjective sparse feature learning model based on the autoencoder. The parameters of the model are learnt by optimizing two objectives, reconstruction error and the sparsity of hidden units simultaneously to find a reasonable compromise between them automatically. We design a multiobjective induced learning procedure for this model based on a multiobjective evolutionary algorithm. In the experiments, we demonstrate that the learning procedure is effective, and the proposed multiobjective model can learn useful sparse features.

Fault detection and classification in electrical power transmission system using artificial neural network.

Science.gov (United States)

Jamil, Majid; Sharma, Sanjeev Kumar; Singh, Rajveer

2015-01-01

This paper focuses on the detection and classification of the faults on electrical power transmission line using artificial neural networks. The three phase currents and voltages of one end are taken as inputs in the proposed scheme. The feed forward neural network along with back propagation algorithm has been employed for detection and classification of the fault for analysis of each of the three phases involved in the process. A detailed analysis with varying number of hidden layers has been performed to validate the choice of the neural network. The simulation results concluded that the present method based on the neural network is efficient in detecting and classifying the faults on transmission lines with satisfactory performances. The different faults are simulated with different parameters to check the versatility of the method. The proposed method can be extended to the Distribution network of the Power System. The various simulations and analysis of signals is done in the MATLAB(®) environment.

Gauging hidden symmetries in two dimensions

International Nuclear Information System (INIS)

Samtleben, Henning; Weidner, Martin

2007-01-01

We initiate the systematic construction of gauged matter-coupled supergravity theories in two dimensions. Subgroups of the affine global symmetry group of toroidally compactified supergravity can be gauged by coupling vector fields with minimal couplings and a particular topological term. The gauge groups typically include hidden symmetries that are not among the target-space isometries of the ungauged theory. The gaugings constructed in this paper are described group-theoretically in terms of a constant embedding tensor subject to a number of constraints which parametrizes the different theories and entirely encodes the gauged Lagrangian. The prime example is the bosonic sector of the maximally supersymmetric theory whose ungauged version admits an affine e 9 global symmetry algebra. The various parameters (related to higher-dimensional p-form fluxes, geometric and non-geometric fluxes, etc.) which characterize the possible gaugings, combine into an embedding tensor transforming in the basic representation of e 9 . This yields an infinite-dimensional class of maximally supersymmetric theories in two dimensions. We work out and discuss several examples of higher-dimensional origin which can be systematically analyzed using the different gradings of e 9

Conjugate descent formulation of backpropagation error in feedforward neural networks

Directory of Open Access Journals (Sweden)

NK Sharma

2009-06-01

Full Text Available The feedforward neural network architecture uses backpropagation learning to determine optimal weights between different interconnected layers. This learning procedure uses a gradient descent technique applied to a sum-of-squares error function for the given input-output pattern. It employs an iterative procedure to minimise the error function for a given set of patterns, by adjusting the weights of the network. The first derivates of the error with respect to the weights identify the local error surface in the descent direction. Hence the network exhibits a different local error surface for every different pattern presented to it, and weights are iteratively modified in order to minimise the current local error. The determination of an optimal weight vector is possible only when the total minimum error (mean of the minimum local errors for all patterns from the training set may be minimised. In this paper, we present a general mathematical formulation for the second derivative of the error function with respect to the weights (which represents a conjugate descent for arbitrary feedforward neural network topologies, and we use this derivative information to obtain the optimal weight vector. The local error is backpropagated among the units of hidden layers via the second order derivative of the error with respect to the weights of the hidden and output layers independently and also in combination. The new total minimum error point may be evaluated with the help of the current total minimum error and the current minimised local error. The weight modification processes is performed twice: once with respect to the present local error and once more with respect to the current total or mean error. We present some numerical evidence that our proposed method yields better network weights than those determined via a conventional gradient descent approach.

A Multilayer Hidden Markov Models-Based Method for Human-Robot Interaction

Directory of Open Access Journals (Sweden)

Chongben Tao

2013-01-01

Full Text Available To achieve Human-Robot Interaction (HRI by using gestures, a continuous gesture recognition approach based on Multilayer Hidden Markov Models (MHMMs is proposed, which consists of two parts. One part is gesture spotting and segment module, the other part is continuous gesture recognition module. Firstly, a Kinect sensor is used to capture 3D acceleration and 3D angular velocity data of hand gestures. And then, a Feed-forward Neural Networks (FNNs and a threshold criterion are used for gesture spotting and segment, respectively. Afterwards, the segmented gesture signals are respectively preprocessed and vector symbolized by a sliding window and a K-means clustering method. Finally, symbolized data are sent into Lower Hidden Markov Models (LHMMs to identify individual gestures, and then, a Bayesian filter with sequential constraints among gestures in Upper Hidden Markov Models (UHMMs is used to correct recognition errors created in LHMMs. Five predefined gestures are used to interact with a Kinect mobile robot in experiments. The experimental results show that the proposed method not only has good effectiveness and accuracy, but also has favorable real-time performance.

Diagonal recurrent neural network based adaptive control of nonlinear dynamical systems using lyapunov stability criterion.

Science.gov (United States)

Kumar, Rajesh; Srivastava, Smriti; Gupta, J R P

2017-03-01

In this paper adaptive control of nonlinear dynamical systems using diagonal recurrent neural network (DRNN) is proposed. The structure of DRNN is a modification of fully connected recurrent neural network (FCRNN). Presence of self-recurrent neurons in the hidden layer of DRNN gives it an ability to capture the dynamic behaviour of the nonlinear plant under consideration (to be controlled). To ensure stability, update rules are developed using lyapunov stability criterion. These rules are then used for adjusting the various parameters of DRNN. The responses of plants obtained with DRNN are compared with those obtained when multi-layer feed forward neural network (MLFFNN) is used as a controller. Also, in example 4, FCRNN is also investigated and compared with DRNN and MLFFNN. Robustness of the proposed control scheme is also tested against parameter variations and disturbance signals. Four simulation examples including one-link robotic manipulator and inverted pendulum are considered on which the proposed controller is applied. The results so obtained show the superiority of DRNN over MLFFNN as a controller. Copyright © 2017 ISA. Published by Elsevier Ltd. All rights reserved.

Artificial neural network modeling and optimization of ultrahigh pressure extraction of green tea polyphenols.

Science.gov (United States)

Xi, Jun; Xue, Yujing; Xu, Yinxiang; Shen, Yuhong

2013-11-01

In this study, the ultrahigh pressure extraction of green tea polyphenols was modeled and optimized by a three-layer artificial neural network. A feed-forward neural network trained with an error back-propagation algorithm was used to evaluate the effects of pressure, liquid/solid ratio and ethanol concentration on the total phenolic content of green tea extracts. The neural network coupled with genetic algorithms was also used to optimize the conditions needed to obtain the highest yield of tea polyphenols. The obtained optimal architecture of artificial neural network model involved a feed-forward neural network with three input neurons, one hidden layer with eight neurons and one output layer including single neuron. The trained network gave the minimum value in the MSE of 0.03 and the maximum value in the R(2) of 0.9571, which implied a good agreement between the predicted value and the actual value, and confirmed a good generalization of the network. Based on the combination of neural network and genetic algorithms, the optimum extraction conditions for the highest yield of green tea polyphenols were determined as follows: 498.8 MPa for pressure, 20.8 mL/g for liquid/solid ratio and 53.6% for ethanol concentration. The total phenolic content of the actual measurement under the optimum predicated extraction conditions was 582.4 ± 0.63 mg/g DW, which was well matched with the predicted value (597.2mg/g DW). This suggests that the artificial neural network model described in this work is an efficient quantitative tool to predict the extraction efficiency of green tea polyphenols. Crown Copyright © 2013. Published by Elsevier Ltd. All rights reserved.

The Dissolved Oxygen Prediction Method Based on Neural Network

Directory of Open Access Journals (Sweden)

Zhong Xiao

2017-01-01

Full Text Available The dissolved oxygen (DO is oxygen dissolved in water, which is an important factor for the aquaculture. Using BP neural network method with the combination of purelin, logsig, and tansig activation functions is proposed for the prediction of aquaculture’s dissolved oxygen. The input layer, hidden layer, and output layer are introduced in detail including the weight adjustment process. The breeding data of three ponds in actual 10 consecutive days were used for experiments; these ponds were located in Beihai, Guangxi, a traditional aquaculture base in southern China. The data of the first 7 days are used for training, and the data of the latter 3 days are used for the test. Compared with the common prediction models, curve fitting (CF, autoregression (AR, grey model (GM, and support vector machines (SVM, the experimental results show that the prediction accuracy of the neural network is the highest, and all the predicted values are less than 5% of the error limit, which can meet the needs of practical applications, followed by AR, GM, SVM, and CF. The prediction model can help to improve the water quality monitoring level of aquaculture which will prevent the deterioration of water quality and the outbreak of disease.

Deep neural network convolution (NNC) for three-class classification of diffuse lung disease opacities in high-resolution CT (HRCT): consolidation, ground-glass opacity (GGO), and normal opacity

Science.gov (United States)

Hashimoto, Noriaki; Suzuki, Kenji; Liu, Junchi; Hirano, Yasushi; MacMahon, Heber; Kido, Shoji

2018-02-01

Consolidation and ground-glass opacity (GGO) are two major types of opacities associated with diffuse lung diseases. Accurate detection and classification of such opacities are crucially important in the diagnosis of lung diseases, but the process is subjective, and suffers from interobserver variability. Our study purpose was to develop a deep neural network convolution (NNC) system for distinguishing among consolidation, GGO, and normal lung tissue in high-resolution CT (HRCT). We developed ensemble of two deep NNC models, each of which was composed of neural network regression (NNR) with an input layer, a convolution layer, a fully-connected hidden layer, and a fully-connected output layer followed by a thresholding layer. The output layer of each NNC provided a map for the likelihood of being each corresponding lung opacity of interest. The two NNC models in the ensemble were connected in a class-selection layer. We trained our NNC ensemble with pairs of input 2D axial slices and "teaching" probability maps for the corresponding lung opacity, which were obtained by combining three radiologists' annotations. We randomly selected 10 and 40 slices from HRCT scans of 172 patients for each class as a training and test set, respectively. Our NNC ensemble achieved an area under the receiver-operating-characteristic (ROC) curve (AUC) of 0.981 and 0.958 in distinction of consolidation and GGO, respectively, from normal opacity, yielding a classification accuracy of 93.3% among 3 classes. Thus, our deep-NNC-based system for classifying diffuse lung diseases achieved high accuracies for classification of consolidation, GGO, and normal opacity.

Prediction of the binary density of the ILs+ water using back-propagated feed forward artificial neural network

Directory of Open Access Journals (Sweden)

Shojaee Safar Ali

2014-01-01

Full Text Available In this study, feasibility of a back-propagated artificial neural network to correlate the binary density of ionic liquids (ILs mixtures containing water as the common solvent has been investigated. To verify the optimized parameters of the neural network, total of 1668 data were collected and divided into two different subsets. The first subsets consisted of more than two-third (1251 data points of data bank was used to find the optimum parameters including weights and biases, number of neurons (7 neurons, transfer functions in hidden and output layer which were tansig and purelin, respectively. In addition, the correlative capability of network was examined using testing subset (417 data points not considered during the training stage. The overall obtained results revealed that the proposed network is accurate enough to correlate the binary density of the ionic liquids mixtures with average absolute relative deviation (AARD % and average relative deviation (ARD % of 1.56% and -0.04 %, respectively. Finally, the correlative capability of the proposed ANN model was compared with one of the available correlations proposed by Rodríguez and Brennecke.

Strong convective storm nowcasting using a hybrid approach of convolutional neural network and hidden Markov model

Science.gov (United States)

Zhang, Wei; Jiang, Ling; Han, Lei

2018-04-01

Convective storm nowcasting refers to the prediction of the convective weather initiation, development, and decay in a very short term (typically 0 2 h) .Despite marked progress over the past years, severe convective storm nowcasting still remains a challenge. With the boom of machine learning, it has been well applied in various fields, especially convolutional neural network (CNN). In this paper, we build a servere convective weather nowcasting system based on CNN and hidden Markov model (HMM) using reanalysis meteorological data. The goal of convective storm nowcasting is to predict if there is a convective storm in 30min. In this paper, we compress the VDRAS reanalysis data to low-dimensional data by CNN as the observation vector of HMM, then obtain the development trend of strong convective weather in the form of time series. It shows that, our method can extract robust features without any artificial selection of features, and can capture the development trend of strong convective storm.

Two-dimensional hidden semantic information model for target saliency detection and eyetracking identification

Science.gov (United States)

Wan, Weibing; Yuan, Lingfeng; Zhao, Qunfei; Fang, Tao

2018-01-01

Saliency detection has been applied to the target acquisition case. This paper proposes a two-dimensional hidden Markov model (2D-HMM) that exploits the hidden semantic information of an image to detect its salient regions. A spatial pyramid histogram of oriented gradient descriptors is used to extract features. After encoding the image by a learned dictionary, the 2D-Viterbi algorithm is applied to infer the saliency map. This model can predict fixation of the targets and further creates robust and effective depictions of the targets' change in posture and viewpoint. To validate the model with a human visual search mechanism, two eyetrack experiments are employed to train our model directly from eye movement data. The results show that our model achieves better performance than visual attention. Moreover, it indicates the plausibility of utilizing visual track data to identify targets.

Synthesis of a novel adaptive wavelet optimized neural cascaded steam blow-off control system for a nuclear power plant

International Nuclear Information System (INIS)

Malik, A.H.; Memon, A.A.; Arshad, F.

2013-01-01

Blow-Off System Controller (MIMO AWNN-SBOSC) is designed based on real time dynamic parametric plant data of steam blow-off system with conventional Single-Input Multi-Output Proportional plus Integral plus Derivative Controller (SIMO PIDC). The proposed MIMO AWANN-SBOSC is designed using three Multi-Input Single-Output Adaptive Wavelet Neural Network based Steam Blow-Off System Controllers (MISO AWNN-SBOSC). The hidden layer of each MISO AWNN-SBOSC is formulated using Mother Wavelet Transforms (MWT). Using nonlinear dynamic neural data of designed MIMO AWNN-SBOSC, a Multi-Input Multi-Output Adaptive Wavelet Neural Network based Steam Blow-Off System Model (MIMO AWNN-SBOSM) is developed in cascaded mode. MIMO AWNN-SBOSM is designed using two MISO AWNN-SBOSM. All training, testing and validation of MIMO AWNN-SBOSC and MIMO AWNN-SBOSM are carried out in MA TLAB while all simulation experiments are performed in Visual C. The results of the new design is evaluated against conventional controller based measured data and found robust, fast and much better in performance. (author)

Methods for discriminating gas-liquid two phase flow patterns based on gray neural networks and SVM

International Nuclear Information System (INIS)

Li Jingjing; Zhou Tao; Duan Jun; Zhang Lei

2013-01-01

Background: The flow patterns of two phase flow will directly influence the heat transfer and mass transfer of the flow. Purpose: By wavelet analysis of the pressure drop experimental data, the wavelet coefficients of different frequency can be obtained. Methods: Get the wavelet energy and then train them in the model of BP neural network to distinguish the flow patterns. Introduced the implant gray neural networks model and use it for the two phase flow for the first time. At the same time, set up the method of training the pressure data and wavelet energy data in the support vector machine. Results: Through treatment of the gray layer, the result of the neural network is more accuracy. It can obviously reduce the effect of data marginalization. The accuracy of the pressure drop Lib-SVM method is 95.2%. Conclusions: The results show that these three methods can make a distinction among the different flow patterns and the Lib-SVM method gets the best result, then the gray neural networks, and at last the BP neural networks. (authors)

Deep neural network with weight sparsity control and pre-training extracts hierarchical features and enhances classification performance: Evidence from whole-brain resting-state functional connectivity patterns of schizophrenia.

Science.gov (United States)

Kim, Junghoe; Calhoun, Vince D; Shim, Eunsoo; Lee, Jong-Hwan

2016-01-01

Functional connectivity (FC) patterns obtained from resting-state functional magnetic resonance imaging data are commonly employed to study neuropsychiatric conditions by using pattern classifiers such as the support vector machine (SVM). Meanwhile, a deep neural network (DNN) with multiple hidden layers has shown its ability to systematically extract lower-to-higher level information of image and speech data from lower-to-higher hidden layers, markedly enhancing classification accuracy. The objective of this study was to adopt the DNN for whole-brain resting-state FC pattern classification of schizophrenia (SZ) patients vs. healthy controls (HCs) and identification of aberrant FC patterns associated with SZ. We hypothesized that the lower-to-higher level features learned via the DNN would significantly enhance the classification accuracy, and proposed an adaptive learning algorithm to explicitly control the weight sparsity in each hidden layer via L1-norm regularization. Furthermore, the weights were initialized via stacked autoencoder based pre-training to further improve the classification performance. Classification accuracy was systematically evaluated as a function of (1) the number of hidden layers/nodes, (2) the use of L1-norm regularization, (3) the use of the pre-training, (4) the use of framewise displacement (FD) removal, and (5) the use of anatomical/functional parcellation. Using FC patterns from anatomically parcellated regions without FD removal, an error rate of 14.2% was achieved by employing three hidden layers and 50 hidden nodes with both L1-norm regularization and pre-training, which was substantially lower than the error rate from the SVM (22.3%). Moreover, the trained DNN weights (i.e., the learned features) were found to represent the hierarchical organization of aberrant FC patterns in SZ compared with HC. Specifically, pairs of nodes extracted from the lower hidden layer represented sparse FC patterns implicated in SZ, which was

Differences between otolith- and semicircular canal-activated neural circuitry in the vestibular system.

Science.gov (United States)

Uchino, Yoshio; Kushiro, Keisuke

2011-12-01

In the last two decades, we have focused on establishing a reliable technique for focal stimulation of vestibular receptors to evaluate neural connectivity. Here, we summarize the vestibular-related neuronal circuits for the vestibulo-ocular reflex, vestibulocollic reflex, and vestibulospinal reflex arcs. The focal stimulating technique also uncovered some hidden neural mechanisms. In the otolith system, we identified two hidden neural mechanisms that enhance otolith receptor sensitivity. The first is commissural inhibition, which boosts sensitivity by incorporating inputs from bilateral otolith receptors, the existence of which was in contradiction to the classical understanding of the otolith system but was observed in the utricular system. The second mechanism, cross-striolar inhibition, intensifies the sensitivity of inputs from both sides of receptive cells across the striola in a single otolith sensor. This was an entirely novel finding and is typically observed in the saccular system. We discuss the possible functional meaning of commissural and cross-striolar inhibition. Finally, our focal stimulating technique was applied to elucidate the different constructions of axonal projections from each vestibular receptor to the spinal cord. We also discuss the possible function of the unique neural connectivity observed in each vestibular receptor system. Copyright © 2011 Elsevier Ireland Ltd and the Japan Neuroscience Society. All rights reserved.

Identification of Non-Linear Structures using Recurrent Neural Networks

DEFF Research Database (Denmark)

Kirkegaard, Poul Henning; Nielsen, Søren R. K.; Hansen, H. I.

Two different partially recurrent neural networks structured as Multi Layer Perceptrons (MLP) are investigated for time domain identification of a non-linear structure.......Two different partially recurrent neural networks structured as Multi Layer Perceptrons (MLP) are investigated for time domain identification of a non-linear structure....

Identification of Non-Linear Structures using Recurrent Neural Networks

DEFF Research Database (Denmark)

Kirkegaard, Poul Henning; Nielsen, Søren R. K.; Hansen, H. I.

1995-01-01

Two different partially recurrent neural networks structured as Multi Layer Perceptrons (MLP) are investigated for time domain identification of a non-linear structure.......Two different partially recurrent neural networks structured as Multi Layer Perceptrons (MLP) are investigated for time domain identification of a non-linear structure....

A review and analysis of neural networks for classification of remotely sensed multispectral imagery

Science.gov (United States)

Paola, Justin D.; Schowengerdt, Robert A.

1993-01-01

A literature survey and analysis of the use of neural networks for the classification of remotely sensed multispectral imagery is presented. As part of a brief mathematical review, the backpropagation algorithm, which is the most common method of training multi-layer networks, is discussed with an emphasis on its application to pattern recognition. The analysis is divided into five aspects of neural network classification: (1) input data preprocessing, structure, and encoding; (2) output encoding and extraction of classes; (3) network architecture, (4) training algorithms; and (5) comparisons to conventional classifiers. The advantages of the neural network method over traditional classifiers are its non-parametric nature, arbitrary decision boundary capabilities, easy adaptation to different types of data and input structures, fuzzy output values that can enhance classification, and good generalization for use with multiple images. The disadvantages of the method are slow training time, inconsistent results due to random initial weights, and the requirement of obscure initialization values (e.g., learning rate and hidden layer size). Possible techniques for ameliorating these problems are discussed. It is concluded that, although the neural network method has several unique capabilities, it will become a useful tool in remote sensing only if it is made faster, more predictable, and easier to use.

Prediction of the rejection of organic compounds (neutral and ionic) by nanofiltration and reverse osmosis membranes using neural networks

Energy Technology Data Exchange (ETDEWEB)

Ammi, Yamina; Khaouane, Latifa; Hanini, Salah [University of Medea, Medea (Algeria)

2015-11-15

This work investigates the use of neural networks in modeling the rejection processes of organic compounds (neutral and ionic) by nanofiltration and reverse osmosis membranes. Three feed-forward neural network (NN) models, characterized by a similar structure (eleven neurons for NN1 and NN2 and twelve neurons for NN3 in the input layer, one hidden layer and one neuron in the output layer), are constructed with the aim of predicting the rejection of organic compounds (neutral and ionic). A set of 956 data points for NN1 and 701 data points for NN2 and NN3 were used to test the neural networks. 80%, 10%, and 10% of the total data were used, respectively, for the training, the validation, and the test of the three models. For the most promising neural network models, the predicted rejection values of the test dataset were compared to measured rejections values; good correlations were found (R= 0.9128 for NN1, R=0.9419 for NN2, and R=0.9527 for NN3). The root mean squared errors for the total dataset were 11.2430% for NN1, 9.0742% for NN2, and 8.2047% for NN3. Furthermore, the comparison between the predicted results and QSAR models shows that the neural network models gave far better.

Prediction of the rejection of organic compounds (neutral and ionic) by nanofiltration and reverse osmosis membranes using neural networks

International Nuclear Information System (INIS)

Ammi, Yamina; Khaouane, Latifa; Hanini, Salah

2015-01-01

This work investigates the use of neural networks in modeling the rejection processes of organic compounds (neutral and ionic) by nanofiltration and reverse osmosis membranes. Three feed-forward neural network (NN) models, characterized by a similar structure (eleven neurons for NN1 and NN2 and twelve neurons for NN3 in the input layer, one hidden layer and one neuron in the output layer), are constructed with the aim of predicting the rejection of organic compounds (neutral and ionic). A set of 956 data points for NN1 and 701 data points for NN2 and NN3 were used to test the neural networks. 80%, 10%, and 10% of the total data were used, respectively, for the training, the validation, and the test of the three models. For the most promising neural network models, the predicted rejection values of the test dataset were compared to measured rejections values; good correlations were found (R= 0.9128 for NN1, R=0.9419 for NN2, and R=0.9527 for NN3). The root mean squared errors for the total dataset were 11.2430% for NN1, 9.0742% for NN2, and 8.2047% for NN3. Furthermore, the comparison between the predicted results and QSAR models shows that the neural network models gave far better.

Neural network models for biological waste-gas treatment systems.

Science.gov (United States)

Rene, Eldon R; Estefanía López, M; Veiga, María C; Kennes, Christian

2011-12-15

This paper outlines the procedure for developing artificial neural network (ANN) based models for three bioreactor configurations used for waste-gas treatment. The three bioreactor configurations chosen for this modelling work were: biofilter (BF), continuous stirred tank bioreactor (CSTB) and monolith bioreactor (MB). Using styrene as the model pollutant, this paper also serves as a general database of information pertaining to the bioreactor operation and important factors affecting gas-phase styrene removal in these biological systems. Biological waste-gas treatment systems are considered to be both advantageous and economically effective in treating a stream of polluted air containing low to moderate concentrations of the target contaminant, over a rather wide range of gas-flow rates. The bioreactors were inoculated with the fungus Sporothrix variecibatus, and their performances were evaluated at different empty bed residence times (EBRT), and at different inlet styrene concentrations (C(i)). The experimental data from these bioreactors were modelled to predict the bioreactors performance in terms of their removal efficiency (RE, %), by adequate training and testing of a three-layered back propagation neural network (input layer-hidden layer-output layer). Two models (BIOF1 and BIOF2) were developed for the BF with different combinations of easily measurable BF parameters as the inputs, that is concentration (gm(-3)), unit flow (h(-1)) and pressure drop (cm of H(2)O). The model developed for the CSTB used two inputs (concentration and unit flow), while the model for the MB had three inputs (concentration, G/L (gas/liquid) ratio, and pressure drop). Sensitivity analysis in the form of absolute average sensitivity (AAS) was performed for all the developed ANN models to ascertain the importance of the different input parameters, and to assess their direct effect on the bioreactors performance. The performance of the models was estimated by the regression

Flank wears Simulation by using back propagation neural network when cutting hardened H-13 steel in CNC End Milling

Science.gov (United States)

Hazza, Muataz Hazza F. Al; Adesta, Erry Y. T.; Riza, Muhammad

2013-12-01

High speed milling has many advantages such as higher removal rate and high productivity. However, higher cutting speed increase the flank wear rate and thus reducing the cutting tool life. Therefore estimating and predicting the flank wear length in early stages reduces the risk of unaccepted tooling cost. This research presents a neural network model for predicting and simulating the flank wear in the CNC end milling process. A set of sparse experimental data for finish end milling on AISI H13 at hardness of 48 HRC have been conducted to measure the flank wear length. Then the measured data have been used to train the developed neural network model. Artificial neural network (ANN) was applied to predict the flank wear length. The neural network contains twenty hidden layer with feed forward back propagation hierarchical. The neural network has been designed with MATLAB Neural Network Toolbox. The results show a high correlation between the predicted and the observed flank wear which indicates the validity of the models.

Flank wears Simulation by using back propagation neural network when cutting hardened H-13 steel in CNC End Milling

International Nuclear Information System (INIS)

Al Hazza, Muataz Hazza F; Adesta, Erry Y T; Riza, Muhammad

2013-01-01

High speed milling has many advantages such as higher removal rate and high productivity. However, higher cutting speed increase the flank wear rate and thus reducing the cutting tool life. Therefore estimating and predicting the flank wear length in early stages reduces the risk of unaccepted tooling cost. This research presents a neural network model for predicting and simulating the flank wear in the CNC end milling process. A set of sparse experimental data for finish end milling on AISI H13 at hardness of 48 HRC have been conducted to measure the flank wear length. Then the measured data have been used to train the developed neural network model. Artificial neural network (ANN) was applied to predict the flank wear length. The neural network contains twenty hidden layer with feed forward back propagation hierarchical. The neural network has been designed with MATLAB Neural Network Toolbox. The results show a high correlation between the predicted and the observed flank wear which indicates the validity of the models

Sentiment analysis: a comparison of deep learning neural network algorithm with SVM and naϊve Bayes for Indonesian text

Science.gov (United States)

Calvin Frans Mariel, Wahyu; Mariyah, Siti; Pramana, Setia

2018-03-01

Deep learning is a new era of machine learning techniques that essentially imitate the structure and function of the human brain. It is a development of deeper Artificial Neural Network (ANN) that uses more than one hidden layer. Deep Learning Neural Network has a great ability on recognizing patterns from various data types such as picture, audio, text, and many more. In this paper, the authors tries to measure that algorithm’s ability by applying it into the text classification. The classification task herein is done by considering the content of sentiment in a text which is also called as sentiment analysis. By using several combinations of text preprocessing and feature extraction techniques, we aim to compare the precise modelling results of Deep Learning Neural Network with the other two commonly used algorithms, the Naϊve Bayes and Support Vector Machine (SVM). This algorithm comparison uses Indonesian text data with balanced and unbalanced sentiment composition. Based on the experimental simulation, Deep Learning Neural Network clearly outperforms the Naϊve Bayes and SVM and offers a better F-1 Score while for the best feature extraction technique which improves that modelling result is Bigram.

Model of Cholera Forecasting Using Artificial Neural Network in Chabahar City, Iran

Directory of Open Access Journals (Sweden)

Zahra Pezeshki

2016-02-01

Full Text Available Background: Cholera as an endemic disease remains a health issue in Iran despite decrease in incidence. Since forecasting epidemic diseases provides appropriate preventive actions in disease spread, different forecasting methods including artificial neural networks have been developed to study parameters involved in incidence and spread of epidemic diseases such as cholera. Objectives: In this study, cholera in rural area of Chabahar, Iran was investigated to achieve a proper forecasting model. Materials and Methods: Data of cholera was gathered from 465 villages, of which 104 reported cholera during ten years period of study. Logistic regression modeling and correlate bivariate were used to determine risk factors and achieve possible predictive model one-hidden-layer perception neural network with backpropagation training algorithm and the sigmoid activation function was trained and tested between the two groups of infected and non-infected villages after preprocessing. For determining validity of prediction, the ROC diagram was used. The study variables included climate conditions and geographical parameters. Results: After determining significant variables of cholera incidence, the described artificial neural network model was capable of forecasting cholera event among villages of test group with accuracy up to 80%. The highest accuracy was achieved when model was trained with variables that were significant in statistical analysis describing that the two methods confirm the result of each other. Conclusions: Application of artificial neural networking assists forecasting cholera for adopting protective measures. For a more accurate prediction, comprehensive information is required including data on hygienic, social and demographic parameters.

A two particle hidden sector and the oscillations with photons

Energy Technology Data Exchange (ETDEWEB)

Alvarez, Pedro D. [Universidad de Antofagasta, Departamento de Fisica, Antofagasta (Chile); Arias, Paola; Maldonado, Carlos [Universidad de Santiago de Chile, Departmento de Fisica, Santiago (Chile)

2018-01-15

We present a detailed study of the oscillations and optical properties for vacuum, in a model for the dark sector that contains axion-like particles and hidden photons. We provide bounds for the couplings versus the mass, using current results from ALPS-I and PVLAS. We also discuss the challenges for the detection of models with more than one hidden particle in light shining trough wall-like experiments. (orig.)

A gentle introduction to artificial neural networks.

Science.gov (United States)

Zhang, Zhongheng

2016-10-01

Artificial neural network (ANN) is a flexible and powerful machine learning technique. However, it is under utilized in clinical medicine because of its technical challenges. The article introduces some basic ideas behind ANN and shows how to build ANN using R in a step-by-step framework. In topology and function, ANN is in analogue to the human brain. There are input and output signals transmitting from input to output nodes. Input signals are weighted before reaching output nodes according to their respective importance. Then the combined signal is processed by activation function. I simulated a simple example to illustrate how to build a simple ANN model using nnet() function. This function allows for one hidden layer with varying number of units in that layer. The basic structure of ANN can be visualized with plug-in plot.nnet() function. The plot function is powerful that it allows for varieties of adjustment to the appearance of the neural networks. Prediction with ANN can be performed with predict() function, similar to that of conventional generalized linear models. Finally, the prediction power of ANN is examined using confusion matrix and average accuracy. It appears that ANN is slightly better than conventional linear model.

The application of artificial neural networks to TLD dose algorithm

International Nuclear Information System (INIS)

Moscovitch, M.

1997-01-01

We review the application of feed forward neural networks to multi element thermoluminescence dosimetry (TLD) dose algorithm development. A Neural Network is an information processing method inspired by the biological nervous system. A dose algorithm based on a neural network is a fundamentally different approach from conventional algorithms, as it has the capability to learn from its own experience. The neural network algorithm is shown the expected dose values (output) associated with a given response of a multi-element dosimeter (input) many times.The algorithm, being trained that way, eventually is able to produce its own unique solution to similar (but not exactly the same) dose calculation problems. For personnel dosimetry, the output consists of the desired dose components: deep dose, shallow dose, and eye dose. The input consists of the TL data obtained from the readout of a multi-element dosimeter. For this application, a neural network architecture was developed based on the concept of functional links network (FLN). The FLN concept allowed an increase in the dimensionality of the input space and construction of a neural network without any hidden layers. This simplifies the problem and results in a relatively simple and reliable dose calculation algorithm. Overall, the neural network dose algorithm approach has been shown to significantly improve the precision and accuracy of dose calculations. (authors)

Refrigerant flow through electronic expansion valve: Experiment and neural network modeling

International Nuclear Information System (INIS)

Cao, Xiang; Li, Ze-Yu; Shao, Liang-Liang; Zhang, Chun-Lu

2016-01-01

Highlights: • Experimental data from different sources were used in comparison of EEV models. • Artificial neural network in EEV modeling is superior to literature correlations. • Artificial neural network with 4-4-1 structure and S function is recommended. • Artificial neural network is flexible for EEV mass flow rate and opening prediction. - Abstract: Electronic expansion valve (EEV) plays a crucial role in controlling refrigerant mass flow rate of refrigeration or heat pump systems for energy savings. However, complexities in two-phase throttling process and geometry make accurate modeling of EEV flow characteristics more difficult. This paper developed an artificial neural network (ANN) model using refrigerant inlet and outlet pressures, inlet subcooling, EEV opening as ANN inputs, refrigerant mass flow rate as ANN output. Both linear and nonlinear transfer functions in hidden layer were used and compared to each other. Experimental data from multiple sources including in-house experiments of one EEV with R410A were used for ANN training and test. In addition, literature correlations were compared with ANN as well. Results showed that the ANN model with nonlinear transfer function worked well in all cases and it is much accurate than the literature correlations. In all cases, nonlinear ANN predicted refrigerant mass flow rates within ±0.4% average relative deviation (A.D.) and 2.7% standard deviation (S.D.), meanwhile it predicted the EEV opening at 0.1% A.D. and 2.1% S.D.

Basic problems and solution methods for two-dimensional continuous 3 × 3 order hidden Markov model

International Nuclear Information System (INIS)

Wang, Guo-gang; Tang, Gui-jin; Gan, Zong-liang; Cui, Zi-guan; Zhu, Xiu-chang

2016-01-01

A novel model referred to as two-dimensional continuous 3 × 3 order hidden Markov model is put forward to avoid the disadvantages of the classical hypothesis of two-dimensional continuous hidden Markov model. This paper presents three equivalent definitions of the model, in which the state transition probability relies on not only immediate horizontal and vertical states but also immediate diagonal state, and in which the probability density of the observation relies on not only current state but also immediate horizontal and vertical states. The paper focuses on the three basic problems of the model, namely probability density calculation, parameters estimation and path backtracking. Some algorithms solving the questions are theoretically derived, by exploiting the idea that the sequences of states on rows or columns of the model can be viewed as states of a one-dimensional continuous 1 × 2 order hidden Markov model. Simulation results further demonstrate the performance of the algorithms. Because there are more statistical characteristics in the structure of the proposed new model, it can more accurately describe some practical problems, as compared to two-dimensional continuous hidden Markov model.

POVMs and hidden variables

International Nuclear Information System (INIS)

Stairs, Allen

2007-01-01

Recent results by Paul Busch and Adan Cabello claim to show that by appealing to POVMs, non-contextual hidden variables can be ruled out in two dimensions. While the results of Busch and Cabello are mathematically correct, interpretive problems render them problematic as no hidden variable proofs

Neural Network-Based Receiver in Band-Limited Communication System with MPPSK Modulation

Directory of Open Access Journals (Sweden)

Wang Zixin

2018-01-01

Full Text Available As a type of the spectrally efficient modulation, the m-ary phase position shift keying (MPPSK has been considered to meet the increasing spectrum requirement in the future wireless system. To limit the signal bandwidth and cancel the out-band interference the band-pass filters are used, which introduce the waveform distortion and inter-symbol interference (ISI. Therefore, a single hidden-layer neural network (NN-based receiver is proposed to jointly equalize and demodulate the received signal. The impulse response of the system is static and the network parameters can be obtained after off-line training. The number of the hidden nodes is also determined through simulations. Simulation results show that the NN-based receiver works well in the communication system with different allocated bandwidths. By observing the modified confusion matrix, the false symbol decision is relevant to modulation index, waveform distortions and the ISI.

Optimization of Melatonin Dissolution from Extended Release Matrices Using Artificial Neural Networking.

Science.gov (United States)

Martarelli, D; Casettari, L; Shalaby, K S; Soliman, M E; Cespi, M; Bonacucina, G; Fagioli, L; Perinelli, D R; Lam, J K W; Palmieri, G F

2016-01-01

Efficacy of melatonin in treating sleep disorders has been demonstrated in numerous studies. Being with short half-life, melatonin needs to be formulated in extended-release tablets to prevent the fast drop of its plasma concentration. However, an attempt to mimic melatonin natural plasma levels during night time is challenging. In this work, Artificial Neural Networks (ANNs) were used to optimize melatonin release from hydrophilic polymer matrices. Twenty-seven different tablet formulations with different amounts of hydroxypropyl methylcellulose, xanthan gum and Carbopol®974P NF were prepared and subjected to drug release studies. Using dissolution test data as inputs for ANN designed by Visual Basic programming language, the ideal number of neurons in the hidden layer was determined trial and error methodology to guarantee the best performance of constructed ANN. Results showed that the ANN with nine neurons in the hidden layer had the best results. ANN was examined to check its predictability and then used to determine the best formula that can mimic the release of melatonin from a marketed brand using similarity fit factor. This work shows the possibility of using ANN to optimize the composition of prolonged-release melatonin tablets having dissolution profile desired.

Hidden Markov models: the best models for forager movements?

Science.gov (United States)

Joo, Rocio; Bertrand, Sophie; Tam, Jorge; Fablet, Ronan

2013-01-01

One major challenge in the emerging field of movement ecology is the inference of behavioural modes from movement patterns. This has been mainly addressed through Hidden Markov models (HMMs). We propose here to evaluate two sets of alternative and state-of-the-art modelling approaches. First, we consider hidden semi-Markov models (HSMMs). They may better represent the behavioural dynamics of foragers since they explicitly model the duration of the behavioural modes. Second, we consider discriminative models which state the inference of behavioural modes as a classification issue, and may take better advantage of multivariate and non linear combinations of movement pattern descriptors. For this work, we use a dataset of >200 trips from human foragers, Peruvian fishermen targeting anchovy. Their movements were recorded through a Vessel Monitoring System (∼1 record per hour), while their behavioural modes (fishing, searching and cruising) were reported by on-board observers. We compare the efficiency of hidden Markov, hidden semi-Markov, and three discriminative models (random forests, artificial neural networks and support vector machines) for inferring the fishermen behavioural modes, using a cross-validation procedure. HSMMs show the highest accuracy (80%), significantly outperforming HMMs and discriminative models. Simulations show that data with higher temporal resolution, HSMMs reach nearly 100% of accuracy. Our results demonstrate to what extent the sequential nature of movement is critical for accurately inferring behavioural modes from a trajectory and we strongly recommend the use of HSMMs for such purpose. In addition, this work opens perspectives on the use of hybrid HSMM-discriminative models, where a discriminative setting for the observation process of HSMMs could greatly improve inference performance.

Hidden Markov models: the best models for forager movements?

Directory of Open Access Journals (Sweden)

Rocio Joo

Full Text Available One major challenge in the emerging field of movement ecology is the inference of behavioural modes from movement patterns. This has been mainly addressed through Hidden Markov models (HMMs. We propose here to evaluate two sets of alternative and state-of-the-art modelling approaches. First, we consider hidden semi-Markov models (HSMMs. They may better represent the behavioural dynamics of foragers since they explicitly model the duration of the behavioural modes. Second, we consider discriminative models which state the inference of behavioural modes as a classification issue, and may take better advantage of multivariate and non linear combinations of movement pattern descriptors. For this work, we use a dataset of >200 trips from human foragers, Peruvian fishermen targeting anchovy. Their movements were recorded through a Vessel Monitoring System (∼1 record per hour, while their behavioural modes (fishing, searching and cruising were reported by on-board observers. We compare the efficiency of hidden Markov, hidden semi-Markov, and three discriminative models (random forests, artificial neural networks and support vector machines for inferring the fishermen behavioural modes, using a cross-validation procedure. HSMMs show the highest accuracy (80%, significantly outperforming HMMs and discriminative models. Simulations show that data with higher temporal resolution, HSMMs reach nearly 100% of accuracy. Our results demonstrate to what extent the sequential nature of movement is critical for accurately inferring behavioural modes from a trajectory and we strongly recommend the use of HSMMs for such purpose. In addition, this work opens perspectives on the use of hybrid HSMM-discriminative models, where a discriminative setting for the observation process of HSMMs could greatly improve inference performance.

A Squeezed Artificial Neural Network for the Symbolic Network Reliability Functions of Binary-State Networks.

Science.gov (United States)

Yeh, Wei-Chang

Network reliability is an important index to the provision of useful information for decision support in the modern world. There is always a need to calculate symbolic network reliability functions (SNRFs) due to dynamic and rapid changes in network parameters. In this brief, the proposed squeezed artificial neural network (SqANN) approach uses the Monte Carlo simulation to estimate the corresponding reliability of a given designed matrix from the Box-Behnken design, and then the Taguchi method is implemented to find the appropriate number of neurons and activation functions of the hidden layer and the output layer in ANN to evaluate SNRFs. According to the experimental results of the benchmark networks, the comparison appears to support the superiority of the proposed SqANN method over the traditional ANN-based approach with at least 16.6% improvement in the median absolute deviation in the cost of extra 2 s on average for all experiments.Network reliability is an important index to the provision of useful information for decision support in the modern world. There is always a need to calculate symbolic network reliability functions (SNRFs) due to dynamic and rapid changes in network parameters. In this brief, the proposed squeezed artificial neural network (SqANN) approach uses the Monte Carlo simulation to estimate the corresponding reliability of a given designed matrix from the Box-Behnken design, and then the Taguchi method is implemented to find the appropriate number of neurons and activation functions of the hidden layer and the output layer in ANN to evaluate SNRFs. According to the experimental results of the benchmark networks, the comparison appears to support the superiority of the proposed SqANN method over the traditional ANN-based approach with at least 16.6% improvement in the median absolute deviation in the cost of extra 2 s on average for all experiments.

Estimating wheat and maize daily evapotranspiration using artificial neural network

Science.gov (United States)

Abrishami, Nazanin; Sepaskhah, Ali Reza; Shahrokhnia, Mohammad Hossein

2018-02-01

In this research, artificial neural network (ANN) is used for estimating wheat and maize daily standard evapotranspiration. Ten ANN models with different structures were designed for each crop. Daily climatic data [maximum temperature (T max), minimum temperature (T min), average temperature (T ave), maximum relative humidity (RHmax), minimum relative humidity (RHmin), average relative humidity (RHave), wind speed (U 2), sunshine hours (n), net radiation (Rn)], leaf area index (LAI), and plant height (h) were used as inputs. For five structures of ten, the evapotranspiration (ETC) values calculated by ETC = ET0 × K C equation (ET0 from Penman-Monteith equation and K C from FAO-56, ANNC) were used as outputs, and for the other five structures, the ETC values measured by weighing lysimeter (ANNM) were used as outputs. In all structures, a feed forward multiple-layer network with one or two hidden layers and sigmoid transfer function and BR or LM training algorithm was used. Favorite network was selected based on various statistical criteria. The results showed the suitable capability and acceptable accuracy of ANNs, particularly those having two hidden layers in their structure in estimating the daily evapotranspiration. Best model for estimation of maize daily evapotranspiration is «M»ANN1 C (8-4-2-1), with T max, T min, RHmax, RHmin, U 2, n, LAI, and h as input data and LM training rule and its statistical parameters (NRMSE, d, and R2) are 0.178, 0.980, and 0.982, respectively. Best model for estimation of wheat daily evapotranspiration is «W»ANN5 C (5-2-3-1), with T max, T min, Rn, LAI, and h as input data and LM training rule, its statistical parameters (NRMSE, d, and R 2) are 0.108, 0.987, and 0.981 respectively. In addition, if the calculated ETC used as the output of the network for both wheat and maize, higher accurate estimation was obtained. Therefore, ANN is suitable method for estimating evapotranspiration of wheat and maize.

Development of Artificial Neural Network Model of Crude Oil Distillation Column

Directory of Open Access Journals (Sweden)

Ali Hussein Khalaf

2016-02-01

Full Text Available Artificial neural network in MATLAB simulator is used to model Baiji crude oil distillation unit based on data generated from aspen-HYSYS simulator. Thirteen inputs, six outputs and over 1487 data set are used to model the actual unit. Nonlinear autoregressive network with exogenous inputs (NARXand back propagation algorithm are used for training. Seventy percent of data are used for training the network while the remaining  thirty percent are used for testing  and validating the network to determine its prediction accuracy. One hidden layer and 34 hidden neurons are used for the proposed network with MSE of 0.25 is obtained. The number of neuron are selected based on less MSE for the network. The model founded to predict the optimal operating conditions for different objective functions within the training limit since ANN models are poor extrapolators. They are usually only reliable within the range of data that they had been trained for.

Development of Artificial Neural Network Model of Crude Oil Distillation Column

Directory of Open Access Journals (Sweden)

Duraid F. Ahmed

2016-02-01

Full Text Available Artificial neural network in MATLAB simulator is used to model Baiji crude oil distillation unit based on data generated from aspen-HYSYS simulator. Thirteen inputs, six outputs and over 1487 data set are used to model the actual unit. Nonlinear autoregressive network with exogenous inputs (NARX and back propagation algorithm are used for training. Seventy percent of data are used for training the network while the remaining thirty percent are used for testing and validating the network to determine its prediction accuracy. One hidden layer and 34 hidden neurons are used for the proposed network with MSE of 0.25 is obtained. The number of neuron are selected based on less MSE for the network. The model founded to predict the optimal operating conditions for different objective functions within the training limit since ANN models are poor extrapolators. They are usually only reliable within the range of data that they had been trained for.

Artificial neural network-aided image analysis system for cell counting.

Science.gov (United States)

Sjöström, P J; Frydel, B R; Wahlberg, L U

1999-05-01

In histological preparations containing debris and synthetic materials, it is difficult to automate cell counting using standard image analysis tools, i.e., systems that rely on boundary contours, histogram thresholding, etc. In an attempt to mimic manual cell recognition, an automated cell counter was constructed using a combination of artificial intelligence and standard image analysis methods. Artificial neural network (ANN) methods were applied on digitized microscopy fields without pre-ANN feature extraction. A three-layer feed-forward network with extensive weight sharing in the first hidden layer was employed and trained on 1,830 examples using the error back-propagation algorithm on a Power Macintosh 7300/180 desktop computer. The optimal number of hidden neurons was determined and the trained system was validated by comparison with blinded human counts. System performance at 50x and lO0x magnification was evaluated. The correlation index at 100x magnification neared person-to-person variability, while 50x magnification was not useful. The system was approximately six times faster than an experienced human. ANN-based automated cell counting in noisy histological preparations is feasible. Consistent histology and computer power are crucial for system performance. The system provides several benefits, such as speed of analysis and consistency, and frees up personnel for other tasks.

Using the Artificial Neural Networks for Forecasting the Risk of Bankruptcy of Banks

Directory of Open Access Journals (Sweden)

Markov Mykhailo Ye.

2018-01-01

Full Text Available The article is aimed at finding the optimal structure of artificial neural network to solve the problem of forecasting the bankruptcy of banks and researching the efficiency of use of the neural networks model for the realities of Ukrainian banking sphere. Results of the research testify that the best accuracy of forecasts for 1-1,5 years showed the model on the basis of the multilayer perceptron with 10 and 2 neurons in the hidden layers. The developed neural networks model can be used as an alternative to statistical methods, as it has shown better results. Prospect for further research in this direction is development of a complex system of support for decision-making for banking institutions, which would include forecasting risks for bank, analysis of the bank’s financial condition and identification of financial problems using innovation instruments and technologies, ensuring the monitoring and control of risks of banking institution. The developed neural networks model can become one of elements of the complex system.

Comments on ‘Temporal significant wave height estimation from wind speed by perceptron Kalman filtering’ by A. Altunkaynak and M. Ozger, Ocean Engineering, Vol. 31(10); 2004,1245-1255

Digital Repository Service at National Institute of Oceanography (India)

Mandal, S.

wind speed. Interestingly the PKF model is a two layered network (input and output) without hidden layer. Also it is a fact that numerical or physical models have restrictions by certain assumptions and conditions, whereas artificial neural network... is shown by Tsai et al (2002). They have carried out forecasting of significant wave heights and periods at a desired location directly from the observed wave records using a supervised artificial neural network with error back-propagation procedures...

Learning to Automatically Detect Features for Mobile Robots Using Second-Order Hidden Markov Models

Directory of Open Access Journals (Sweden)

Olivier Aycard

2004-12-01

Full Text Available In this paper, we propose a new method based on Hidden Markov Models to interpret temporal sequences of sensor data from mobile robots to automatically detect features. Hidden Markov Models have been used for a long time in pattern recognition, especially in speech recognition. Their main advantages over other methods (such as neural networks are their ability to model noisy temporal signals of variable length. We show in this paper that this approach is well suited for interpretation of temporal sequences of mobile-robot sensor data. We present two distinct experiments and results: the first one in an indoor environment where a mobile robot learns to detect features like open doors or T-intersections, the second one in an outdoor environment where a different mobile robot has to identify situations like climbing a hill or crossing a rock.

A One-Layer Recurrent Neural Network for Real-Time Portfolio Optimization With Probability Criterion.

Science.gov (United States)

Liu, Qingshan; Dang, Chuangyin; Huang, Tingwen

2013-02-01

This paper presents a decision-making model described by a recurrent neural network for dynamic portfolio optimization. The portfolio-optimization problem is first converted into a constrained fractional programming problem. Since the objective function in the programming problem is not convex, the traditional optimization techniques are no longer applicable for solving this problem. Fortunately, the objective function in the fractional programming is pseudoconvex on the feasible region. It leads to a one-layer recurrent neural network modeled by means of a discontinuous dynamic system. To ensure the optimal solutions for portfolio optimization, the convergence of the proposed neural network is analyzed and proved. In fact, the neural network guarantees to get the optimal solutions for portfolio-investment advice if some mild conditions are satisfied. A numerical example with simulation results substantiates the effectiveness and illustrates the characteristics of the proposed neural network.

Two layer powder pressing

International Nuclear Information System (INIS)

Schreiner, H.

1979-01-01

First, significance and advantages of sintered materials consisting of two layers are pointed out. By means of the two layer powder pressing technique metal powders are formed resulting in compacts with high accuracy of shape and mass. Attributes of basic powders, different filling methods and pressing techniques are discussed. The described technique is supposed to find further applications in the field of two layer compacts in the near future

Fast learning method for convolutional neural networks using extreme learning machine and its application to lane detection.

Science.gov (United States)

Kim, Jihun; Kim, Jonghong; Jang, Gil-Jin; Lee, Minho

2017-03-01

Deep learning has received significant attention recently as a promising solution to many problems in the area of artificial intelligence. Among several deep learning architectures, convolutional neural networks (CNNs) demonstrate superior performance when compared to other machine learning methods in the applications of object detection and recognition. We use a CNN for image enhancement and the detection of driving lanes on motorways. In general, the process of lane detection consists of edge extraction and line detection. A CNN can be used to enhance the input images before lane detection by excluding noise and obstacles that are irrelevant to the edge detection result. However, training conventional CNNs requires considerable computation and a big dataset. Therefore, we suggest a new learning algorithm for CNNs using an extreme learning machine (ELM). The ELM is a fast learning method used to calculate network weights between output and hidden layers in a single iteration and thus, can dramatically reduce learning time while producing accurate results with minimal training data. A conventional ELM can be applied to networks with a single hidden layer; as such, we propose a stacked ELM architecture in the CNN framework. Further, we modify the backpropagation algorithm to find the targets of hidden layers and effectively learn network weights while maintaining performance. Experimental results confirm that the proposed method is effective in reducing learning time and improving performance. Copyright © 2016 Elsevier Ltd. All rights reserved.

Forcast of TEXT plasma disruptions using soft X-rays as input signal in a neural network

International Nuclear Information System (INIS)

Vannucci, A.; Oliveira, K.A.; Tajima, T.

1998-02-01

A feed-forward neural network with two hidden layers is used in this work to forecast major and minor disruptive instabilities in TEXT discharges. Using soft X-ray signals as input data, the neural net is trained with one disruptive plasma pulse, and a different disruptive discharge is used for validation. After being properly trained the networks, with the same set of weights. is then used to forecast disruptions in two others different plasma pulses. It is observed that the neural net is able to predict the incoming of a disruption more than 3 ms in advance. This time interval is almost three times longer than the one already obtained previously when magnetic signal from a Mirnov coil was used to feed the neural networks with. To our own eye we fail to see any indication of an upcoming disruption from the experimental data this far back from the time of disruption. Finally, from what we observe in the predictive behavior of our network, speculations are made whether the disruption triggering mechanism would be associated to an increase of the m = 2 magnetic island, that disturbs the central part of the plasma column afterwards or, in face of the results from this work, the initial perturbation would have occurred first in the central part of the plasma column, within the q = 1 magnetic surface, and then the m = 2 MHD mode would be destabilized afterwards

Multistability in bidirectional associative memory neural networks

International Nuclear Information System (INIS)

Huang Gan; Cao Jinde

2008-01-01

In this Letter, the multistability issue is studied for Bidirectional Associative Memory (BAM) neural networks. Based on the existence and stability analysis of the neural networks with or without delay, it is found that the 2n-dimensional networks can have 3 n equilibria and 2 n equilibria of them are locally exponentially stable, where each layer of the BAM network has n neurons. Furthermore, the results has been extended to (n+m)-dimensional BAM neural networks, where there are n and m neurons on the two layers respectively. Finally, two numerical examples are presented to illustrate the validity of our results

Multistability in bidirectional associative memory neural networks

Science.gov (United States)

Huang, Gan; Cao, Jinde

2008-04-01

In this Letter, the multistability issue is studied for Bidirectional Associative Memory (BAM) neural networks. Based on the existence and stability analysis of the neural networks with or without delay, it is found that the 2 n-dimensional networks can have 3 equilibria and 2 equilibria of them are locally exponentially stable, where each layer of the BAM network has n neurons. Furthermore, the results has been extended to (n+m)-dimensional BAM neural networks, where there are n and m neurons on the two layers respectively. Finally, two numerical examples are presented to illustrate the validity of our results.

Artificial Neural Networks to Predict the Power Output of a PV Panel

Directory of Open Access Journals (Sweden)

Valerio Lo Brano

2014-01-01

Full Text Available The paper illustrates an adaptive approach based on different topologies of artificial neural networks (ANNs for the power energy output forecasting of photovoltaic (PV modules. The analysis of the PV module’s power output needed detailed local climate data, which was collected by a dedicated weather monitoring system. The Department of Energy, Information Engineering, and Mathematical Models of the University of Palermo (Italy has built up a weather monitoring system that worked together with a data acquisition system. The power output forecast is obtained using three different types of ANNs: a one hidden layer Multilayer perceptron (MLP, a recursive neural network (RNN, and a gamma memory (GM trained with the back propagation. In order to investigate the influence of climate variability on the electricity production, the ANNs were trained using weather data (air temperature, solar irradiance, and wind speed along with historical power output data available for the two test modules. The model validation was performed by comparing model predictions with power output data that were not used for the network's training. The results obtained bear out the suitability of the adopted methodology for the short-term power output forecasting problem and identified the best topology.

Neural networks applied to discriminate botanical origin of honeys.

Science.gov (United States)

Anjos, Ofélia; Iglesias, Carla; Peres, Fátima; Martínez, Javier; García, Ángela; Taboada, Javier

2015-05-15

The aim of this work is develop a tool based on neural networks to predict the botanical origin of honeys using physical and chemical parameters. The managed database consists of 49 honey samples of 2 different classes: monofloral (almond, holm oak, sweet chestnut, eucalyptus, orange, rosemary, lavender, strawberry trees, thyme, heather, sunflower) and multifloral. The moisture content, electrical conductivity, water activity, ashes content, pH, free acidity, colorimetric coordinates in CIELAB space (L(∗), a(∗), b(∗)) and total phenols content of the honey samples were evaluated. Those properties were considered as input variables of the predictive model. The neural network is optimised through several tests with different numbers of neurons in the hidden layer and also with different input variables. The reduced error rates (5%) allow us to conclude that the botanical origin of honey can be reliably and quickly known from the colorimetric information and the electrical conductivity of honey. Copyright © 2014 Elsevier Ltd. All rights reserved.

An implementation of Elman neural network for polycystic ovary classification based on ultrasound images

Science.gov (United States)

Thufailah, I. F.; Adiwijaya; Wisesty, U. N.; Jondri

2018-03-01

Polycystic Ovary Syndrome (PCOS) is a reproduction problem that causes irregular menstruation period. Insulin and androgen hormone have big roles for this problem. This syndrome should be detected shortly, since it is able to cause a more serious disease, such as cardiovascular, diabetes, and obesity. The detection of this syndrome is done by analyzing ovary morphology and hormone test. However, the more economical way of test is by identifying the ovary morphology using ultrasonography. To classify whether one ovary is normal or it has polycystic ovary (PCO) follicle, the analysis will be done manually by a gynecologist. This paper will design a system to detect PCO using Gabor Wavelet method for feature extraction and Elman Neural Network is used to classify PCO and non-PCO. Elman Neural Network is chosen because it contains context layer to recall the previous condition. This paper compared the accuracy and process time of each dataset, then also did testing on elman’s parameters, such as layer delay, hidden layer, and training function. Based on tests done in this paper, the most accurate number is 78.1% with 32 features.

Familial or Sporadic Idiopathic Scoliosis – classification based on artificial neural network and GAPDH and ACTB transcription profile

Science.gov (United States)

2013-01-01

Background Importance of hereditary factors in the etiology of Idiopathic Scoliosis is widely accepted. In clinical practice some of the IS patients present with positive familial history of the deformity and some do not. Traditionally about 90% of patients have been considered as sporadic cases without familial recurrence. However the exact proportion of Familial and Sporadic Idiopathic Scoliosis is still unknown. Housekeeping genes encode proteins that are usually essential for the maintenance of basic cellular functions. ACTB and GAPDH are two housekeeping genes encoding respectively a cytoskeletal protein β-actin, and glyceraldehyde-3-phosphate dehydrogenase, an enzyme of glycolysis. Although their expression levels can fluctuate between different tissues and persons, human housekeeping genes seem to exhibit a preserved tissue-wide expression ranking order. It was hypothesized that expression ranking order of two representative housekeeping genes ACTB and GAPDH might be disturbed in the tissues of patients with Familial Idiopathic Scoliosis (with positive family history of idiopathic scoliosis) opposed to the patients with no family members affected (Sporadic Idiopathic Scoliosis). An artificial neural network (ANN) was developed that could serve to differentiate between familial and sporadic cases of idiopathic scoliosis based on the expression levels of ACTB and GAPDH in different tissues of scoliotic patients. The aim of the study was to investigate whether the expression levels of ACTB and GAPDH in different tissues of idiopathic scoliosis patients could be used as a source of data for specially developed artificial neural network in order to predict the positive family history of index patient. Results The comparison of developed models showed, that the most satisfactory classification accuracy was achieved for ANN model with 18 nodes in the first hidden layer and 16 nodes in the second hidden layer. The classification accuracy for positive Idiopathic

Data Hiding Based on a Two-Layer Turtle Shell Matrix

Directory of Open Access Journals (Sweden)

Xiao-Zhu Xie

2018-02-01

Full Text Available Data hiding is a technology that embeds data into a cover carrier in an imperceptible way while still allowing the hidden data to be extracted accurately from the stego-carrier, which is one important branch of computer science and has drawn attention of scholars in the last decade. Turtle shell-based (TSB schemes have become popular in recent years due to their higher embedding capacity (EC and better visual quality of the stego-image than most of the none magic matrices based (MMB schemes. This paper proposes a two-layer turtle shell matrix-based (TTSMB scheme for data hiding, in which an extra attribute presented by a 4-ary digit is assigned to each element of the turtle shell matrix with symmetrical distribution. Therefore, compared with the original TSB scheme, two more bits are embedded into each pixel pair to obtain a higher EC up to 2.5 bits per pixel (bpp. The experimental results reveal that under the condition of the same visual quality, the EC of the proposed scheme outperforms state-of-the-art data hiding schemes.

Ant colony optimization and neural networks applied to nuclear power plant monitoring

International Nuclear Information System (INIS)

Santos, Gean Ribeiro dos; Andrade, Delvonei Alves de; Pereira, Iraci Martinez

2015-01-01

A recurring challenge in production processes is the development of monitoring and diagnosis systems. Those systems help on detecting unexpected changes and interruptions, preventing losses and mitigating risks. Artificial Neural Networks (ANNs) have been extensively used in creating monitoring systems. Usually the ANNs created to solve this kind of problem are created by taking into account only parameters as the number of inputs, outputs, and hidden layers. The result networks are generally fully connected and have no improvements in its topology. This work intends to use an Ant Colony Optimization (ACO) algorithm to create a tuned neural network. The ACO search algorithm will use Back Error Propagation (BP) to optimize the network topology by suggesting the best neuron connections. The result ANN will be applied to monitoring the IEA-R1 research reactor at IPEN. (author)

Ant colony optimization and neural networks applied to nuclear power plant monitoring

Energy Technology Data Exchange (ETDEWEB)

Santos, Gean Ribeiro dos; Andrade, Delvonei Alves de; Pereira, Iraci Martinez, E-mail: gean@usp.br, E-mail: delvonei@ipen.br, E-mail: martinez@ipen.br [Instituto de Pesquisas Energeticas e Nucleares (IPEN/CNEN-SP), Sao Paulo, SP (Brazil)

2015-07-01

A recurring challenge in production processes is the development of monitoring and diagnosis systems. Those systems help on detecting unexpected changes and interruptions, preventing losses and mitigating risks. Artificial Neural Networks (ANNs) have been extensively used in creating monitoring systems. Usually the ANNs created to solve this kind of problem are created by taking into account only parameters as the number of inputs, outputs, and hidden layers. The result networks are generally fully connected and have no improvements in its topology. This work intends to use an Ant Colony Optimization (ACO) algorithm to create a tuned neural network. The ACO search algorithm will use Back Error Propagation (BP) to optimize the network topology by suggesting the best neuron connections. The result ANN will be applied to monitoring the IEA-R1 research reactor at IPEN. (author)

A fuzzy neural network model to forecast the percent cloud coverage and cloud top temperature maps

Directory of Open Access Journals (Sweden)

Y. Tulunay

2008-12-01

Full Text Available Atmospheric processes are highly nonlinear. A small group at the METU in Ankara has been working on a fuzzy data driven generic model of nonlinear processes. The model developed is called the Middle East Technical University Fuzzy Neural Network Model (METU-FNN-M. The METU-FNN-M consists of a Fuzzy Inference System (METU-FIS, a data driven Neural Network module (METU-FNN of one hidden layer and several neurons, and a mapping module, which employs the Bezier Surface Mapping technique. In this paper, the percent cloud coverage (%CC and cloud top temperatures (CTT are forecast one month ahead of time at 96 grid locations. The probable influence of cosmic rays and sunspot numbers on cloudiness is considered by using the METU-FNN-M.

Beyond Retinal Layers: A Deep Voting Model for Automated Geographic Atrophy Segmentation in SD-OCT Images.

Science.gov (United States)

Ji, Zexuan; Chen, Qiang; Niu, Sijie; Leng, Theodore; Rubin, Daniel L

2018-01-01

To automatically and accurately segment geographic atrophy (GA) in spectral-domain optical coherence tomography (SD-OCT) images by constructing a voting system with deep neural networks without the use of retinal layer segmentation. An automatic GA segmentation method for SD-OCT images based on the deep network was constructed. The structure of the deep network was composed of five layers, including one input layer, three hidden layers, and one output layer. During the training phase, the labeled A-scans with 1024 features were directly fed into the network as the input layer to obtain the deep representations. Then a soft-max classifier was trained to determine the label of each individual pixel. Finally, a voting decision strategy was used to refine the segmentation results among 10 trained models. Two image data sets with GA were used to evaluate the model. For the first dataset, our algorithm obtained a mean overlap ratio (OR) 86.94% ± 8.75%, absolute area difference (AAD) 11.49% ± 11.50%, and correlation coefficients (CC) 0.9857; for the second dataset, the mean OR, AAD, and CC of the proposed method were 81.66% ± 10.93%, 8.30% ± 9.09%, and 0.9952, respectively. The proposed algorithm was capable of improving over 5% and 10% segmentation accuracy, respectively, when compared with several state-of-the-art algorithms on two data sets. Without retinal layer segmentation, the proposed algorithm could produce higher segmentation accuracy and was more stable when compared with state-of-the-art methods that relied on retinal layer segmentation results. Our model may provide reliable GA segmentations from SD-OCT images and be useful in the clinical diagnosis of advanced nonexudative AMD. Based on the deep neural networks, this study presents an accurate GA segmentation method for SD-OCT images without using any retinal layer segmentation results, and may contribute to improved understanding of advanced nonexudative AMD.

A one-layer recurrent neural network for non-smooth convex optimization subject to linear inequality constraints

International Nuclear Information System (INIS)

Liu, Xiaolan; Zhou, Mi

2016-01-01

In this paper, a one-layer recurrent network is proposed for solving a non-smooth convex optimization subject to linear inequality constraints. Compared with the existing neural networks for optimization, the proposed neural network is capable of solving more general convex optimization with linear inequality constraints. The convergence of the state variables of the proposed neural network to achieve solution optimality is guaranteed as long as the designed parameters in the model are larger than the derived lower bounds.

Early tube leak detection system for steam boiler at KEV power plant

Directory of Open Access Journals (Sweden)

Ismail Firas B.

2016-01-01

Full Text Available Tube leakage in boilers has been a major contribution to trips which eventually leads to power plant shut downs. Training of network and developing artificial neural network (ANN models are essential in fault detection in critically large systems. This research focusses on the ANN modelling through training and validation of real data acquired from a sub-critical boiler unit. The artificial neural network (ANN was used to develop a compatible model and to evaluate the working properties and behaviour of boiler. The training and validation of real data has been applied using the feed-forward with back-propagation (BP. The right combination of number of neurons, number of hidden layers, training algorithms and training functions was run to achieve the best ANN model with lowest error. The ANN was trained and validated using real site data acquired from a coal fired power plant in Malaysia. The results showed that the Neural Network (NN with one hidden layers performed better than two hidden layer using feed-forward back-propagation network. The outcome from this study give us the best ANN model which eventually allows for early detection of boiler tube leakages, and forecast of a trip before the real shutdown. This will eventually reduce shutdowns in power plants.

A One-Layer Recurrent Neural Network for Pseudoconvex Optimization Problems With Equality and Inequality Constraints.

Science.gov (United States)

Qin, Sitian; Yang, Xiudong; Xue, Xiaoping; Song, Jiahui

2017-10-01

Pseudoconvex optimization problem, as an important nonconvex optimization problem, plays an important role in scientific and engineering applications. In this paper, a recurrent one-layer neural network is proposed for solving the pseudoconvex optimization problem with equality and inequality constraints. It is proved that from any initial state, the state of the proposed neural network reaches the feasible region in finite time and stays there thereafter. It is also proved that the state of the proposed neural network is convergent to an optimal solution of the related problem. Compared with the related existing recurrent neural networks for the pseudoconvex optimization problems, the proposed neural network in this paper does not need the penalty parameters and has a better convergence. Meanwhile, the proposed neural network is used to solve three nonsmooth optimization problems, and we make some detailed comparisons with the known related conclusions. In the end, some numerical examples are provided to illustrate the effectiveness of the performance of the proposed neural network.

Study on pattern recognition of Raman spectrum based on fuzzy neural network

Science.gov (United States)

Zheng, Xiangxiang; Lv, Xiaoyi; Mo, Jiaqing

2017-10-01

Hydatid disease is a serious parasitic disease in many regions worldwide, especially in Xinjiang, China. Raman spectrum of the serum of patients with echinococcosis was selected as the research object in this paper. The Raman spectrum of blood samples from healthy people and patients with echinococcosis are measured, of which the spectrum characteristics are analyzed. The fuzzy neural network not only has the ability of fuzzy logic to deal with uncertain information, but also has the ability to store knowledge of neural network, so it is combined with the Raman spectrum on the disease diagnosis problem based on Raman spectrum. Firstly, principal component analysis (PCA) is used to extract the principal components of the Raman spectrum, reducing the network input and accelerating the prediction speed and accuracy of Network based on remaining the original data. Then, the information of the extracted principal component is used as the input of the neural network, the hidden layer of the network is the generation of rules and the inference process, and the output layer of the network is fuzzy classification output. Finally, a part of samples are randomly selected for the use of training network, then the trained network is used for predicting the rest of the samples, and the predicted results are compared with general BP neural network to illustrate the feasibility and advantages of fuzzy neural network. Success in this endeavor would be helpful for the research work of spectroscopic diagnosis of disease and it can be applied in practice in many other spectral analysis technique fields.

Neural network-based segmentation of satellite imagery for estimating house cluster of an urban settlement from Google Earth images

International Nuclear Information System (INIS)

Wardaya, P D; Ridha, S

2014-01-01

In this paper a backpropagation neural network is utilized to perform house cluster segmentation from Google Earth data. The algorithm is subjected to identify houses in the image based on the RGB pattern within each pixel. Training data is given through cropping selection for a target that is a house cluster and a non object. The algorithm assigns 1 to a pixel belong to a class of object and 0 to a class of non object. The resulting outcome, a binary image, is then utilized to perform quantification to estimate the number of house clusters. The number of the hidden layer is varying in order to find its effect to the neural network performance and total computational time

Hidden photons in connection to dark matter

Energy Technology Data Exchange (ETDEWEB)

Andreas, Sarah; Ringwald, Andreas [Deutsches Elektronen-Synchrotron (DESY), Hamburg (Germany); Goodsell, Mark D. [CPhT, Ecole Polytechnique, Palaiseau (France)

2013-06-15

Light extra U(1) gauge bosons, so called hidden photons, which reside in a hidden sector have attracted much attention since they are a well motivated feature of many scenarios beyond the Standard Model and furthermore could mediate the interaction with hidden sector dark matter.We review limits on hidden photons from past electron beam dump experiments including two new limits from such experiments at KEK and Orsay. In addition, we study the possibility of having dark matter in the hidden sector. A simple toy model and different supersymmetric realisations are shown to provide viable dark matter candidates in the hidden sector that are in agreement with recent direct detection limits.

Hidden photons in connection to dark matter

International Nuclear Information System (INIS)

Andreas, Sarah; Ringwald, Andreas; Goodsell, Mark D.

2013-06-01

Light extra U(1) gauge bosons, so called hidden photons, which reside in a hidden sector have attracted much attention since they are a well motivated feature of many scenarios beyond the Standard Model and furthermore could mediate the interaction with hidden sector dark matter.We review limits on hidden photons from past electron beam dump experiments including two new limits from such experiments at KEK and Orsay. In addition, we study the possibility of having dark matter in the hidden sector. A simple toy model and different supersymmetric realisations are shown to provide viable dark matter candidates in the hidden sector that are in agreement with recent direct detection limits.

Probing many-body localization with neural networks

Science.gov (United States)

Schindler, Frank; Regnault, Nicolas; Neupert, Titus

2017-06-01

We show that a simple artificial neural network trained on entanglement spectra of individual states of a many-body quantum system can be used to determine the transition between a many-body localized and a thermalizing regime. Specifically, we study the Heisenberg spin-1/2 chain in a random external field. We employ a multilayer perceptron with a single hidden layer, which is trained on labeled entanglement spectra pertaining to the fully localized and fully thermal regimes. We then apply this network to classify spectra belonging to states in the transition region. For training, we use a cost function that contains, in addition to the usual error and regularization parts, a term that favors a confident classification of the transition region states. The resulting phase diagram is in good agreement with the one obtained by more conventional methods and can be computed for small systems. In particular, the neural network outperforms conventional methods in classifying individual eigenstates pertaining to a single disorder realization. It allows us to map out the structure of these eigenstates across the transition with spatial resolution. Furthermore, we analyze the network operation using the dreaming technique to show that the neural network correctly learns by itself the power-law structure of the entanglement spectra in the many-body localized regime.

Optimization of multilayer neural network parameters for speaker recognition

Science.gov (United States)

Tovarek, Jaromir; Partila, Pavol; Rozhon, Jan; Voznak, Miroslav; Skapa, Jan; Uhrin, Dominik; Chmelikova, Zdenka

2016-05-01

This article discusses the impact of multilayer neural network parameters for speaker identification. The main task of speaker identification is to find a specific person in the known set of speakers. It means that the voice of an unknown speaker (wanted person) belongs to a group of reference speakers from the voice database. One of the requests was to develop the text-independent system, which means to classify wanted person regardless of content and language. Multilayer neural network has been used for speaker identification in this research. Artificial neural network (ANN) needs to set parameters like activation function of neurons, steepness of activation functions, learning rate, the maximum number of iterations and a number of neurons in the hidden and output layers. ANN accuracy and validation time are directly influenced by the parameter settings. Different roles require different settings. Identification accuracy and ANN validation time were evaluated with the same input data but different parameter settings. The goal was to find parameters for the neural network with the highest precision and shortest validation time. Input data of neural networks are a Mel-frequency cepstral coefficients (MFCC). These parameters describe the properties of the vocal tract. Audio samples were recorded for all speakers in a laboratory environment. Training, testing and validation data set were split into 70, 15 and 15 %. The result of the research described in this article is different parameter setting for the multilayer neural network for four speakers.

Appropriateness of Dropout Layers and Allocation of Their 0.5 Rates across Convolutional Neural Networks for CIFAR-10, EEACL26, and NORB Datasets

OpenAIRE

Romanuke Vadim V.

2017-01-01

A technique of DropOut for preventing overfitting of convolutional neural networks for image classification is considered in the paper. The goal is to find a rule of rationally allocating DropOut layers of 0.5 rate to maximise performance. To achieve the goal, two common network architectures are used having either 4 or 5 convolutional layers. Benchmarking is fulfilled with CIFAR-10, EEACL26, and NORB datasets. Initially, series of all admissible versions for allocation of DropOut layers are ...

Optimization the Initial Weights of Artificial Neural Networks via Genetic Algorithm Applied to Hip Bone Fracture Prediction

Directory of Open Access Journals (Sweden)

Yu-Tzu Chang

2012-01-01

Full Text Available This paper aims to find the optimal set of initial weights to enhance the accuracy of artificial neural networks (ANNs by using genetic algorithms (GA. The sample in this study included 228 patients with first low-trauma hip fracture and 215 patients without hip fracture, both of them were interviewed with 78 questions. We used logistic regression to select 5 important factors (i.e., bone mineral density, experience of fracture, average hand grip strength, intake of coffee, and peak expiratory flow rate for building artificial neural networks to predict the probabilities of hip fractures. Three-layer (one hidden layer ANNs models with back-propagation training algorithms were adopted. The purpose in this paper is to find the optimal initial weights of neural networks via genetic algorithm to improve the predictability. Area under the ROC curve (AUC was used to assess the performance of neural networks. The study results showed the genetic algorithm obtained an AUC of 0.858±0.00493 on modeling data and 0.802 ± 0.03318 on testing data. They were slightly better than the results of our previous study (0.868±0.00387 and 0.796±0.02559, resp.. Thus, the preliminary study for only using simple GA has been proved to be effective for improving the accuracy of artificial neural networks.

Moving image compression and generalization capability of constructive neural networks

Science.gov (United States)

Ma, Liying; Khorasani, Khashayar

2001-03-01

To date numerous techniques have been proposed to compress digital images to ease their storage and transmission over communication channels. Recently, a number of image compression algorithms using Neural Networks NNs have been developed. Particularly, several constructive feed-forward neural networks FNNs have been proposed by researchers for image compression, and promising results have been reported. At the previous SPIE AeroSense conference 2000, we proposed to use a constructive One-Hidden-Layer Feedforward Neural Network OHL-FNN for compressing digital images. In this paper, we first investigate the generalization capability of the proposed OHL-FNN in the presence of additive noise for network training and/ or generalization. Extensive experimental results for different scenarios are presented. It is revealed that the constructive OHL-FNN is not as robust to additive noise in input image as expected. Next, the constructive OHL-FNN is applied to moving images, video sequences. The first, or other specified frame in a moving image sequence is used to train the network. The remaining moving images that follow are then generalized/compressed by this trained network. Three types of correlation-like criteria measuring the similarity of any two images are introduced. The relationship between the generalization capability of the constructed net and the similarity of images is investigated in some detail. It is shown that the constructive OHL-FNN is promising even for changing images such as those extracted from a football game.

An Artificial Neural Networks Approach to Estimate Occupational Accident: A National Perspective for Turkey

Directory of Open Access Journals (Sweden)

Hüseyin Ceylan

2014-01-01

Full Text Available Occupational accident estimation models were developed by using artificial neural networks (ANNs for Turkey. Using these models the number of occupational accidents and death and permanent incapacity numbers resulting from occupational accidents were estimated for Turkey until the year of 2025 by the three different scenarios. In the development of the models, insured workers, workplace, occupational accident, death, and permanent incapacity values were used as model parameters with data between 1970 and 2012. 2-5-1 neural network architecture was selected as the best network architecture. Sigmoid was used in hidden layers and linear function was used at output layer. The feed forward back propagation algorithm was used to train the network. In order to obtain a useful model, the network was trained between 1970 and 1999 to estimate the values of 2000 to 2012. The result was compared with the real values and it was seen that it is applicable for this aim. The performances of all developed models were evaluated using mean absolute percent errors (MAPE, mean absolute errors (MAE, and root mean square errors (RMSE.

A Grey Wolf Optimizer for Modular Granular Neural Networks for Human Recognition

Directory of Open Access Journals (Sweden)

Daniela Sánchez

2017-01-01

Full Text Available A grey wolf optimizer for modular neural network (MNN with a granular approach is proposed. The proposed method performs optimal granulation of data and design of modular neural networks architectures to perform human recognition, and to prove its effectiveness benchmark databases of ear, iris, and face biometric measures are used to perform tests and comparisons against other works. The design of a modular granular neural network (MGNN consists in finding optimal parameters of its architecture; these parameters are the number of subgranules, percentage of data for the training phase, learning algorithm, goal error, number of hidden layers, and their number of neurons. Nowadays, there is a great variety of approaches and new techniques within the evolutionary computing area, and these approaches and techniques have emerged to help find optimal solutions to problems or models and bioinspired algorithms are part of this area. In this work a grey wolf optimizer is proposed for the design of modular granular neural networks, and the results are compared against a genetic algorithm and a firefly algorithm in order to know which of these techniques provides better results when applied to human recognition.

Layers and Multilayers of Self-Assembled Polymers: Tunable Engineered Extracellular Matrix Coatings for Neural Cell Growth.

Science.gov (United States)

Landry, Michael J; Rollet, Frédéric-Guillaume; Kennedy, Timothy E; Barrett, Christopher J

2018-03-12

Growing primary cells and tissue in long-term cultures, such as primary neural cell culture, presents many challenges. A critical component of any environment that supports neural cell growth in vivo is an appropriate 2-D surface or 3-D scaffold, typically in the form of a thin polymer layer that coats an underlying plastic or glass substrate and aims to mimic critical aspects of the extracellular matrix. A fundamental challenge to mimicking a hydrophilic, soft natural cell environment is that materials with these properties are typically fragile and are difficult to adhere to and stabilize on an underlying plastic or glass cell culture substrate. In this review, we highlight the current state of the art and overview recent developments of new artificial extracellular matrix (ECM) surfaces for in vitro neural cell culture. Notably, these materials aim to strike a balance between being hydrophilic and soft while also being thick, stable, robust, and bound well to the underlying surface to provide an effective surface to support long-term cell growth. We focus on improved surface and scaffold coating systems that can mimic the natural physicochemical properties that enhance neuronal survival and growth, applied as soft hydrophilic polymer coatings for both in vitro cell culture and for implantable neural probes and 3-D matrixes that aim to enhance stability and longevity to promote neural biocompatibility in vivo. With respect to future developments, we outline four emerging principles that serve to guide the development of polymer assemblies that function well as artificial ECMs: (a) design inspired by biological systems and (b) the employment of principles of aqueous soft bonding and self-assembly to achieve (c) a high-water-content gel-like coating that is stable over time in a biological environment and possesses (d) a low modulus to more closely mimic soft, compliant real biological tissue. We then highlight two emerging classes of thick material coatings that

Fuzzy Counter Propagation Neural Network Control for a Class of Nonlinear Dynamical Systems.

Science.gov (United States)

Sakhre, Vandana; Jain, Sanjeev; Sapkal, Vilas S; Agarwal, Dev P

2015-01-01

Fuzzy Counter Propagation Neural Network (FCPN) controller design is developed, for a class of nonlinear dynamical systems. In this process, the weight connecting between the instar and outstar, that is, input-hidden and hidden-output layer, respectively, is adjusted by using Fuzzy Competitive Learning (FCL). FCL paradigm adopts the principle of learning, which is used to calculate Best Matched Node (BMN) which is proposed. This strategy offers a robust control of nonlinear dynamical systems. FCPN is compared with the existing network like Dynamic Network (DN) and Back Propagation Network (BPN) on the basis of Mean Absolute Error (MAE), Mean Square Error (MSE), Best Fit Rate (BFR), and so forth. It envisages that the proposed FCPN gives better results than DN and BPN. The effectiveness of the proposed FCPN algorithms is demonstrated through simulations of four nonlinear dynamical systems and multiple input and single output (MISO) and a single input and single output (SISO) gas furnace Box-Jenkins time series data.

Fuzzy Counter Propagation Neural Network Control for a Class of Nonlinear Dynamical Systems

Directory of Open Access Journals (Sweden)

Vandana Sakhre

2015-01-01

Full Text Available Fuzzy Counter Propagation Neural Network (FCPN controller design is developed, for a class of nonlinear dynamical systems. In this process, the weight connecting between the instar and outstar, that is, input-hidden and hidden-output layer, respectively, is adjusted by using Fuzzy Competitive Learning (FCL. FCL paradigm adopts the principle of learning, which is used to calculate Best Matched Node (BMN which is proposed. This strategy offers a robust control of nonlinear dynamical systems. FCPN is compared with the existing network like Dynamic Network (DN and Back Propagation Network (BPN on the basis of Mean Absolute Error (MAE, Mean Square Error (MSE, Best Fit Rate (BFR, and so forth. It envisages that the proposed FCPN gives better results than DN and BPN. The effectiveness of the proposed FCPN algorithms is demonstrated through simulations of four nonlinear dynamical systems and multiple input and single output (MISO and a single input and single output (SISO gas furnace Box-Jenkins time series data.

Diagnosis method utilizing neural networks

International Nuclear Information System (INIS)

Watanabe, K.; Tamayama, K.

1990-01-01

Studies have been made on the technique of neural networks, which will be used to identify a cause of a small anomalous state in the reactor coolant system of the ATR (Advance Thermal Reactor). Three phases of analyses were carried out in this study. First, simulation for 100 seconds was made to determine how the plant parameters respond after the occurence of a transient decrease in reactivity, flow rate and temperature of feed water and increase in the steam flow rate and steam pressure, which would produce a decrease of water level in a steam drum of the ATR. Next, the simulation data was analysed utilizing an autoregressive model. From this analysis, a total of 36 coherency functions up to 0.5 Hz in each transient were computed among nine important and detectable plant parameters: neutron flux, flow rate of coolant, steam or feed water, water level in the steam drum, pressure and opening area of control valve in a steam pipe, feed water temperature and electrical power. Last, learning of neural networks composed of 96 input, 4-9 hidden and 5 output layer units was done by use of the generalized delta rule, namely a back-propagation algorithm. These convergent computations were continued as far as the difference between the desired outputs, 1 for direct cause or 0 for four other ones and actual outputs reached less than 10%. (1) Coherency functions were not governed by decreasing rate of reactivity in the range of 0.41x10 -2 dollar/s to 1.62x10 -2 dollar /s or by decreasing depth of the feed water temperature in the range of 3 deg C to 10 deg C or by a change of 10% or less in the three other causes. Change in coherency functions only depended on the type of cause. (2) The direct cause from the other four ones could be discriminated with 0.94+-0.01 of output level. A maximum of 0.06 output height was found among the other four causes. (3) Calculation load which is represented as products of learning times and numbers of the hidden units did not depend on the

HTTR operation monitoring with neural network in 30 days operation at 850degC

International Nuclear Information System (INIS)

Shimizu, Atsushi; Nabeshima, Kunihiko; Nakagawa, Shigeaki

2009-01-01

The High temperature engineering test reactor (HTTR) executed the rated power operation for 30days of the first time (850degC in temperature of the nuclear reactor outlet coolant) until March, 27th through April, 26th, 2007. In this operation, HTTR was observed according to the operation monitoring model with the neural network, and the detection performance of neural network was verified during slight changes of reactor state at rated power. The neural network used for the operation monitoring was an auto-associative network, where 31 input 31 outputs and the hidden layers were connected with 20 units by the hierarchy of three layer structure. Back-propagation algorithm was used for study rule. The operation monitoring model in initial study was constructed by using the power up data between 30% and rated power, which were randomly studied. The adjustment study during the operation monitoring changes the internal structure of the initial study model to follow the changes of reactor status, such as the burn-up of the nuclear fuel for the rated power operation. As a monitoring result, slight changes of reactor state by the control system operation were correctly detected, and the on-line application to an early anomaly diagnosis for HTTR facilities will be expected. (author)

Hidden Costs of Hospital Based Delivery from Two Tertiary Hospitals in Western Nepal.

Directory of Open Access Journals (Sweden)

Jeevan Acharya

Full Text Available Hospital based delivery has been an expensive experience for poor households because of hidden costs which are usually unaccounted in hospital costs. The main aim of this study was to estimate the hidden costs of hospital based delivery and determine the factors associated with the hidden costs.A hospital based cross-sectional study was conducted among 384 post-partum mothers with their husbands/house heads during the discharge time in Manipal Teaching Hospital and Western Regional Hospital, Pokhara, Nepal. A face to face interview with each respondent was conducted using a structured questionnaire. Hidden costs were calculated based on the price rate of the market during the time of the study.The total hidden costs for normal delivery and C-section delivery were 243.4 USD (US Dollar and 321.6 USD respectively. Of the total maternity care expenditures; higher mean expenditures were found for food & drinking (53.07%, clothes (9.8% and transport (7.3%. For postpartum women with their husband or house head, the total mean opportunity cost of "days of work loss" were 84.1 USD and 81.9 USD for normal delivery and C-section respectively. Factors such as literate mother (p = 0.007, employed house head (p = 0.011, monthly family income more than 25,000 NRs (Nepalese Rupees (p = 0.014, private hospital as a place of delivery (p = 0.0001, C-section as a mode of delivery (p = 0.0001, longer duration (>5days of stay in hospital (p = 0.0001, longer distance (>15km from house to hospital (p = 0.0001 and longer travel time (>240 minutes from house to hospital (p = 0.007 showed a significant association with the higher hidden costs (>25000 NRs.Experiences of hidden costs on hospital based delivery and opportunity costs of days of work loss were found high. Several socio-demographic factors, delivery related factors (place and mode of delivery, length of stay, distance from hospital and travel time were associated with hidden costs. Hidden costs can be a

Hidden Costs of Hospital Based Delivery from Two Tertiary Hospitals in Western Nepal.

Science.gov (United States)

Acharya, Jeevan; Kaehler, Nils; Marahatta, Sujan Babu; Mishra, Shiva Raj; Subedi, Sudarshan; Adhikari, Bipin

2016-01-01

Hospital based delivery has been an expensive experience for poor households because of hidden costs which are usually unaccounted in hospital costs. The main aim of this study was to estimate the hidden costs of hospital based delivery and determine the factors associated with the hidden costs. A hospital based cross-sectional study was conducted among 384 post-partum mothers with their husbands/house heads during the discharge time in Manipal Teaching Hospital and Western Regional Hospital, Pokhara, Nepal. A face to face interview with each respondent was conducted using a structured questionnaire. Hidden costs were calculated based on the price rate of the market during the time of the study. The total hidden costs for normal delivery and C-section delivery were 243.4 USD (US Dollar) and 321.6 USD respectively. Of the total maternity care expenditures; higher mean expenditures were found for food & drinking (53.07%), clothes (9.8%) and transport (7.3%). For postpartum women with their husband or house head, the total mean opportunity cost of "days of work loss" were 84.1 USD and 81.9 USD for normal delivery and C-section respectively. Factors such as literate mother (p = 0.007), employed house head (p = 0.011), monthly family income more than 25,000 NRs (Nepalese Rupees) (p = 0.014), private hospital as a place of delivery (p = 0.0001), C-section as a mode of delivery (p = 0.0001), longer duration (>5days) of stay in hospital (p = 0.0001), longer distance (>15km) from house to hospital (p = 0.0001) and longer travel time (>240 minutes) from house to hospital (p = 0.007) showed a significant association with the higher hidden costs (>25000 NRs). Experiences of hidden costs on hospital based delivery and opportunity costs of days of work loss were found high. Several socio-demographic factors, delivery related factors (place and mode of delivery, length of stay, distance from hospital and travel time) were associated with hidden costs. Hidden costs can be a

3D Polygon Mesh Compression with Multi Layer Feed Forward Neural Networks

Directory of Open Access Journals (Sweden)

Emmanouil Piperakis

2003-06-01

Full Text Available In this paper, an experiment is conducted which proves that multi layer feed forward neural networks are capable of compressing 3D polygon meshes. Our compression method not only preserves the initial accuracy of the represented object but also enhances it. The neural network employed includes the vertex coordinates, the connectivity and normal information in one compact form, converting the discrete and surface polygon representation into an analytic, solid colloquial. Furthermore, the 3D object in its compressed neural form can be directly - without decompression - used for rendering. The neural compression - representation is viable to 3D transformations without the need of any anti-aliasing techniques - transformations do not disrupt the accuracy of the geometry. Our method does not su.er any scaling problem and was tested with objects of 300 to 107 polygons - such as the David of Michelangelo - achieving in all cases an order of O(b3 less bits for the representation than any other commonly known compression method. The simplicity of our algorithm and the established mathematical background of neural networks combined with their aptness for hardware implementation can establish this method as a good solution for polygon compression and if further investigated, a novel approach for 3D collision, animation and morphing.

Exploration of artificial neural network [ANN] to predict the electrochemical characteristics of lithium-ion cells

Energy Technology Data Exchange (ETDEWEB)

Parthiban, Thirumalai; Ravi, R.; Kalaiselvi, N. [Central Electrochemical Research Institute (CECRI), Karaikudi 630006 (India)

2007-12-31

CoO anode, as an alternate to the carbonaceous anodes of lithium-ion cells has been prepared and investigated for electrochemical charge-discharge characteristics for about 50 cycles. Artificial neural networks (ANNs), which are useful in estimating battery performance, has been deployed for the first time to forecast and to verify the charge-discharge behavior of lithium-ion cells containing CoO anode for a total of 50 cycles. In this novel approach, ANN that has one input layer with one neuron corresponding to one input variable, viz., cycles [charge-discharge cycles] and a hidden layer consisting of three neurons to produce their outputs to the output layer through a sigmoid function has been selected for the present investigation. The output layer consists of two neurons, representing the charge and discharge capacity, whose activation function is also the sigmoid transfer function. In this ever first attempt to exploit ANN as an effective theoretical tool to understand the charge-discharge characteristics of lithium-ion cells, an excellent agreement between the calculated and observed capacity values was found with CoO anodes with the best fit values corresponding to an error factor of <1%, which is the highlight of the present study. (author)

Large-N limit of the two-Hermitian-matrix model by the hidden BRST method

International Nuclear Information System (INIS)

Alfaro, J.

1993-01-01

This paper discusses the large-N limit of the two-Hermitian-matrix model in zero dimensions, using the hidden Becchi-Rouet-Stora-Tyutin method. A system of integral equations previously found is solved, showing that it contained the exact solution of the model in leading order of large N

Optimization of artificial neural network models through genetic algorithms for surface ozone concentration forecasting.

Science.gov (United States)

Pires, J C M; Gonçalves, B; Azevedo, F G; Carneiro, A P; Rego, N; Assembleia, A J B; Lima, J F B; Silva, P A; Alves, C; Martins, F G

2012-09-01

This study proposes three methodologies to define artificial neural network models through genetic algorithms (GAs) to predict the next-day hourly average surface ozone (O(3)) concentrations. GAs were applied to define the activation function in hidden layer and the number of hidden neurons. Two of the methodologies define threshold models, which assume that the behaviour of the dependent variable (O(3) concentrations) changes when it enters in a different regime (two and four regimes were considered in this study). The change from one regime to another depends on a specific value (threshold value) of an explanatory variable (threshold variable), which is also defined by GAs. The predictor variables were the hourly average concentrations of carbon monoxide (CO), nitrogen oxide, nitrogen dioxide (NO(2)), and O(3) (recorded in the previous day at an urban site with traffic influence) and also meteorological data (hourly averages of temperature, solar radiation, relative humidity and wind speed). The study was performed for the period from May to August 2004. Several models were achieved and only the best model of each methodology was analysed. In threshold models, the variables selected by GAs to define the O(3) regimes were temperature, CO and NO(2) concentrations, due to their importance in O(3) chemistry in an urban atmosphere. In the prediction of O(3) concentrations, the threshold model that considers two regimes was the one that fitted the data most efficiently.

Population Decoding of Motor Cortical Activity using a Generalized Linear Model with Hidden States

Science.gov (United States)

Lawhern, Vernon; Wu, Wei; Hatsopoulos, Nicholas G.; Paninski, Liam

2010-01-01

Generalized linear models (GLMs) have been developed for modeling and decoding population neuronal spiking activity in the motor cortex. These models provide reasonable characterizations between neural activity and motor behavior. However, they lack a description of movement-related terms which are not observed directly in these experiments, such as muscular activation, the subject's level of attention, and other internal or external states. Here we propose to include a multi-dimensional hidden state to address these states in a GLM framework where the spike count at each time is described as a function of the hand state (position, velocity, and acceleration), truncated spike history, and the hidden state. The model can be identified by an Expectation-Maximization algorithm. We tested this new method in two datasets where spikes were simultaneously recorded using a multi-electrode array in the primary motor cortex of two monkeys. It was found that this method significantly improves the model-fitting over the classical GLM, for hidden dimensions varying from 1 to 4. This method also provides more accurate decoding of hand state (lowering the Mean Square Error by up to 29% in some cases), while retaining real-time computational efficiency. These improvements on representation and decoding over the classical GLM model suggest that this new approach could contribute as a useful tool to motor cortical decoding and prosthetic applications. PMID:20359500

Comparison of logistic regression and artificial neural network in low back pain prediction: second national health survey.

Science.gov (United States)

Parsaeian, M; Mohammad, K; Mahmoudi, M; Zeraati, H

2012-01-01

The purpose of this investigation was to compare empirically predictive ability of an artificial neural network with a logistic regression in prediction of low back pain. Data from the second national health survey were considered in this investigation. This data includes the information of low back pain and its associated risk factors among Iranian people aged 15 years and older. Artificial neural network and logistic regression models were developed using a set of 17294 data and they were validated in a test set of 17295 data. Hosmer and Lemeshow recommendation for model selection was used in fitting the logistic regression. A three-layer perceptron with 9 inputs, 3 hidden and 1 output neurons was employed. The efficiency of two models was compared by receiver operating characteristic analysis, root mean square and -2 Loglikelihood criteria. The area under the ROC curve (SE), root mean square and -2Loglikelihood of the logistic regression was 0.752 (0.004), 0.3832 and 14769.2, respectively. The area under the ROC curve (SE), root mean square and -2Loglikelihood of the artificial neural network was 0.754 (0.004), 0.3770 and 14757.6, respectively. Based on these three criteria, artificial neural network would give better performance than logistic regression. Although, the difference is statistically significant, it does not seem to be clinically significant.

Prediction of geomagnetic storms from solar wind data with the use of a neural network

Directory of Open Access Journals (Sweden)

H. Lundstedt

Full Text Available An artificial feed-forward neural network with one hidden layer and error back-propagation learning is used to predict the geomagnetic activity index (D_st one hour in advance. The B_z-component and Σ_Bz, the density, and the velocity of the solar wind are used as input to the network. The network is trained on data covering a total of 8700 h, extracted from the 25-year period from 1963 to 1987, taken from the NSSDC data base. The performance of the network is examined with test data, not included in the training set, which covers 386 h and includes four different storms. Whilst the network predicts the initial and main phase well, the recovery phase is not modelled correctly, implying that a single hidden layer error back-propagation network is not enough, if the measured D_st is not available instantaneously. The performance of the network is independent of whether the raw parameters are used, or the electric field and square root of the dynamical pressure.

Observation of hidden Fermi surface nesting in a two dimensional conductor

International Nuclear Information System (INIS)

Breuer, K.; Stagerescu, C.; Smith, K.E.; Greenblatt, M.; Ramanujachary, K.

1996-01-01

We report the first direct measurement of hidden Fermi surface nesting in a two dimensional conductor. The system studied was Na 0.9 Mo 6 O 17 , and the measured Fermi surface consists of electron and hole pockets that can be combined to form sets of pseudo-one-dimensional Fermi surfaces, exhibiting the nesting necessary to drive a Peierls transition to a charge density wave state. The observed nesting vector is shown to be in excellent agreement with theory. copyright 1996 The American Physical Society

Rule extraction from minimal neural networks for credit card screening.

Science.gov (United States)

Setiono, Rudy; Baesens, Bart; Mues, Christophe

2011-08-01

While feedforward neural networks have been widely accepted as effective tools for solving classification problems, the issue of finding the best network architecture remains unresolved, particularly so in real-world problem settings. We address this issue in the context of credit card screening, where it is important to not only find a neural network with good predictive performance but also one that facilitates a clear explanation of how it produces its predictions. We show that minimal neural networks with as few as one hidden unit provide good predictive accuracy, while having the added advantage of making it easier to generate concise and comprehensible classification rules for the user. To further reduce model size, a novel approach is suggested in which network connections from the input units to this hidden unit are removed by a very straightaway pruning procedure. In terms of predictive accuracy, both the minimized neural networks and the rule sets generated from them are shown to compare favorably with other neural network based classifiers. The rules generated from the minimized neural networks are concise and thus easier to validate in a real-life setting.

Hypothetical Pattern Recognition Design Using Multi-Layer Perceptorn Neural Network For Supervised Learning

Directory of Open Access Journals (Sweden)

Md. Abdullah-al-mamun

2015-08-01

Full Text Available Abstract Humans are capable to identifying diverse shape in the different pattern in the real world as effortless fashion due to their intelligence is grow since born with facing several learning process. Same way we can prepared an machine using human like brain called Artificial Neural Network that can be recognize different pattern from the real world object. Although the various techniques is exists to implementation the pattern recognition but recently the artificial neural network approaches have been giving the significant attention. Because the approached of artificial neural network is like a human brain that is learn from different observation and give a decision the previously learning rule. Over the 50 years research now a days pattern recognition for machine learning using artificial neural network got a significant achievement. For this reason many real world problem can be solve by modeling the pattern recognition process. The objective of this paper is to present the theoretical concept for pattern recognition design using Multi-Layer Perceptorn neural networkin the algorithm of artificial Intelligence as the best possible way of utilizing available resources to make a decision that can be a human like performance.

Hidden symmetries in N-layer dielectric stacks

Science.gov (United States)

Liu, Haihao; Shoufie Ukhtary, M.; Saito, Riichiro

2017-11-01

The optical properties of a multilayer system with arbitrary N layers of dielectric media are investigated. Each layer is one of two dielectric media, with a thickness one-quarter the wavelength of light in that medium, corresponding to a central frequency f 0. Using the transfer matrix method, the transmittance T is calculated for all possible 2 N sequences for small N. Unexpectedly, it is found that instead of 2 N different values of T at f 0 (T 0), there are only (N/2+1) discrete values of T 0, for even N, and (N + 1) for odd N. We explain this high degeneracy in T 0 values by finding symmetry operations on the sequences that do not change T 0. Analytical formulae were derived for the T 0 values and their degeneracies as functions of N and an integer parameter for each sequence we call ‘charge’. Additionally, the bandwidth at f 0 and filter response of the transmission spectra are investigated, revealing asymptotic behavior at large N.

Inferring topologies via driving-based generalized synchronization of two-layer networks

Science.gov (United States)

Wang, Yingfei; Wu, Xiaoqun; Feng, Hui; Lu, Jun-an; Xu, Yuhua

2016-05-01

The interaction topology among the constituents of a complex network plays a crucial role in the network’s evolutionary mechanisms and functional behaviors. However, some network topologies are usually unknown or uncertain. Meanwhile, coupling delays are ubiquitous in various man-made and natural networks. Hence, it is necessary to gain knowledge of the whole or partial topology of a complex dynamical network by taking into consideration communication delay. In this paper, topology identification of complex dynamical networks is investigated via generalized synchronization of a two-layer network. Particularly, based on the LaSalle-type invariance principle of stochastic differential delay equations, an adaptive control technique is proposed by constructing an auxiliary layer and designing proper control input and updating laws so that the unknown topology can be recovered upon successful generalized synchronization. Numerical simulations are provided to illustrate the effectiveness of the proposed method. The technique provides a certain theoretical basis for topology inference of complex networks. In particular, when the considered network is composed of systems with high-dimension or complicated dynamics, a simpler response layer can be constructed, which is conducive to circuit design. Moreover, it is practical to take into consideration perturbations caused by control input. Finally, the method is applicable to infer topology of a subnetwork embedded within a complex system and locate hidden sources. We hope the results can provide basic insight into further research endeavors on understanding practical and economical topology inference of networks.

A one-layer recurrent neural network for constrained pseudoconvex optimization and its application for dynamic portfolio optimization.

Science.gov (United States)

Liu, Qingshan; Guo, Zhishan; Wang, Jun

2012-02-01

In this paper, a one-layer recurrent neural network is proposed for solving pseudoconvex optimization problems subject to linear equality and bound constraints. Compared with the existing neural networks for optimization (e.g., the projection neural networks), the proposed neural network is capable of solving more general pseudoconvex optimization problems with equality and bound constraints. Moreover, it is capable of solving constrained fractional programming problems as a special case. The convergence of the state variables of the proposed neural network to achieve solution optimality is guaranteed as long as the designed parameters in the model are larger than the derived lower bounds. Numerical examples with simulation results illustrate the effectiveness and characteristics of the proposed neural network. In addition, an application for dynamic portfolio optimization is discussed. Copyright © 2011 Elsevier Ltd. All rights reserved.

Hidden Liquidity

OpenAIRE

Cebiroglu, Gökhan; Horst, Ulrich

2012-01-01

We cross-sectionally analyze the presence of aggregated hidden depth and trade volume in the S&P 500 and identify its key determinants. We find that the spread is the main predictor for a stock’s hidden dimension, both in terms of traded and posted liquidity. Our findings moreover suggest that large hidden orders are associated with larger transaction costs, higher price impact and increased volatility. In particular, as large hidden orders fail to attract (latent) liquidity to the market, hi...

Satisfiability of logic programming based on radial basis function neural networks

International Nuclear Information System (INIS)

Hamadneh, Nawaf; Sathasivam, Saratha; Tilahun, Surafel Luleseged; Choon, Ong Hong

2014-01-01

In this paper, we propose a new technique to test the Satisfiability of propositional logic programming and quantified Boolean formula problem in radial basis function neural networks. For this purpose, we built radial basis function neural networks to represent the proportional logic which has exactly three variables in each clause. We used the Prey-predator algorithm to calculate the output weights of the neural networks, while the K-means clustering algorithm is used to determine the hidden parameters (the centers and the widths). Mean of the sum squared error function is used to measure the activity of the two algorithms. We applied the developed technique with the recurrent radial basis function neural networks to represent the quantified Boolean formulas. The new technique can be applied to solve many applications such as electronic circuits and NP-complete problems

Satisfiability of logic programming based on radial basis function neural networks

Energy Technology Data Exchange (ETDEWEB)

Hamadneh, Nawaf; Sathasivam, Saratha; Tilahun, Surafel Luleseged; Choon, Ong Hong [School of Mathematical Sciences, Universiti Sains Malaysia, 11800 USM, Penang (Malaysia)

2014-07-10

In this paper, we propose a new technique to test the Satisfiability of propositional logic programming and quantified Boolean formula problem in radial basis function neural networks. For this purpose, we built radial basis function neural networks to represent the proportional logic which has exactly three variables in each clause. We used the Prey-predator algorithm to calculate the output weights of the neural networks, while the K-means clustering algorithm is used to determine the hidden parameters (the centers and the widths). Mean of the sum squared error function is used to measure the activity of the two algorithms. We applied the developed technique with the recurrent radial basis function neural networks to represent the quantified Boolean formulas. The new technique can be applied to solve many applications such as electronic circuits and NP-complete problems.

Neural node network and model, and method of teaching same

Science.gov (United States)

Parlos, Alexander G. (Inventor); Atiya, Amir F. (Inventor); Fernandez, Benito (Inventor); Tsai, Wei K. (Inventor); Chong, Kil T. (Inventor)

1995-01-01

The present invention is a fully connected feed forward network that includes at least one hidden layer 16. The hidden layer 16 includes nodes 20 in which the output of the node is fed back to that node as an input with a unit delay produced by a delay device 24 occurring in the feedback path 22 (local feedback). Each node within each layer also receives a delayed output (crosstalk) produced by a delay unit 36 from all the other nodes within the same layer 16. The node performs a transfer function operation based on the inputs from the previous layer and the delayed outputs. The network can be implemented as analog or digital or within a general purpose processor. Two teaching methods can be used: (1) back propagation of weight calculation that includes the local feedback and the crosstalk or (2) more preferably a feed forward gradient decent which immediately follows the output computations and which also includes the local feedback and the crosstalk. Subsequent to the gradient propagation, the weights can be normalized, thereby preventing convergence to a local optimum. Education of the network can be incremental both on and off-line. An educated network is suitable for modeling and controlling dynamic nonlinear systems and time series systems and predicting the outputs as well as hidden states and parameters. The educated network can also be further educated during on-line processing.

Neural architecture design based on extreme learning machine.

Science.gov (United States)

Bueno-Crespo, Andrés; García-Laencina, Pedro J; Sancho-Gómez, José-Luis

2013-12-01

Selection of the optimal neural architecture to solve a pattern classification problem entails to choose the relevant input units, the number of hidden neurons and its corresponding interconnection weights. This problem has been widely studied in many research works but their solutions usually involve excessive computational cost in most of the problems and they do not provide a unique solution. This paper proposes a new technique to efficiently design the MultiLayer Perceptron (MLP) architecture for classification using the Extreme Learning Machine (ELM) algorithm. The proposed method provides a high generalization capability and a unique solution for the architecture design. Moreover, the selected final network only retains those input connections that are relevant for the classification task. Experimental results show these advantages. Copyright © 2013 Elsevier Ltd. All rights reserved.

Artificial neural networks in the evaluation of the radioactive waste drums activity

International Nuclear Information System (INIS)

Potiens, J.R.A.J.; Hiromoto, G.

2006-01-01

The mathematical techniques are becoming more important to solve geometry and standard identification problems. The gamma spectrometry of radioactive waste drums would be a complex solution problem. The main difficulty is the detectors calibration for this geometry; the waste is not homogeneously distributed inside the drums, therefore there are many possible combinations between the activity and the position of these radionuclides inside the drums, making the preparation of calibration standards impracticable. This work describes the development of a methodology to estimate the activity of a 200 L radioactive waste drum, as well as a mapping of the waste distribution, using Artificial Neural Network. The neural network data set entry obtaining was based on the possible detection efficiency combination with 10 sources activities varying from 0 to 74 x 10 3 Bq. The set up consists of a 200 L drum divided in 5 layers. Ten detectors were positioned all the way through a parallel line to the drum axis, from 15 cm of its surface. The Cesium -137 radionuclide source was used. The 50 efficiency obtained values (10 detectors and 5 layers), combined with the 10 source intensities resulted in a 100,000 lines for 15 columns matrix, with all the possible combinations of source intensity and the Cs-137 position in the 5 layers of the drum. This archive was divided in 2 parts to compose the set of training: input and target files. The MatLab 7.0 module of neural networks was used for training. The net architecture has 10 neurons in the input layer, 18 in the hidden layer and 5 in the output layer. The training algorithm was the 'traincgb' and after 300 'epoch s' the medium square error was 0.00108172. This methodology allows knowing the detection positions answers in a heterogeneous distribution of radionuclides inside a 200 L waste drum; in consequence it is possible to estimate the total activity of the drum in the training neural network limits. The results accuracy depends

Artificial Neural Network-Based System for PET Volume Segmentation

Directory of Open Access Journals (Sweden)

Mhd Saeed Sharif

2010-01-01

Full Text Available Tumour detection, classification, and quantification in positron emission tomography (PET imaging at early stage of disease are important issues for clinical diagnosis, assessment of response to treatment, and radiotherapy planning. Many techniques have been proposed for segmenting medical imaging data; however, some of the approaches have poor performance, large inaccuracy, and require substantial computation time for analysing large medical volumes. Artificial intelligence (AI approaches can provide improved accuracy and save decent amount of time. Artificial neural networks (ANNs, as one of the best AI techniques, have the capability to classify and quantify precisely lesions and model the clinical evaluation for a specific problem. This paper presents a novel application of ANNs in the wavelet domain for PET volume segmentation. ANN performance evaluation using different training algorithms in both spatial and wavelet domains with a different number of neurons in the hidden layer is also presented. The best number of neurons in the hidden layer is determined according to the experimental results, which is also stated Levenberg-Marquardt backpropagation training algorithm as the best training approach for the proposed application. The proposed intelligent system results are compared with those obtained using conventional techniques including thresholding and clustering based approaches. Experimental and Monte Carlo simulated PET phantom data sets and clinical PET volumes of nonsmall cell lung cancer patients were utilised to validate the proposed algorithm which has demonstrated promising results.

Compressing the hidden variable space of a qubit

OpenAIRE

Montina, Alberto

2010-01-01

In previously exhibited hidden variable models of quantum state preparation and measurement, the number of continuous hidden variables describing the actual state of a single realization is never smaller than the quantum state manifold dimension. We introduce a simple model for a qubit whose hidden variable space is one-dimensional, i.e., smaller than the two-dimensional Bloch sphere. The hidden variable probability distributions associated with the quantum states satisfy reasonable criteria ...

Bayesian Inference and Online Learning in Poisson Neuronal Networks.

Science.gov (United States)

Huang, Yanping; Rao, Rajesh P N

2016-08-01

Motivated by the growing evidence for Bayesian computation in the brain, we show how a two-layer recurrent network of Poisson neurons can perform both approximate Bayesian inference and learning for any hidden Markov model. The lower-layer sensory neurons receive noisy measurements of hidden world states. The higher-layer neurons infer a posterior distribution over world states via Bayesian inference from inputs generated by sensory neurons. We demonstrate how such a neuronal network with synaptic plasticity can implement a form of Bayesian inference similar to Monte Carlo methods such as particle filtering. Each spike in a higher-layer neuron represents a sample of a particular hidden world state. The spiking activity across the neural population approximates the posterior distribution over hidden states. In this model, variability in spiking is regarded not as a nuisance but as an integral feature that provides the variability necessary for sampling during inference. We demonstrate how the network can learn the likelihood model, as well as the transition probabilities underlying the dynamics, using a Hebbian learning rule. We present results illustrating the ability of the network to perform inference and learning for arbitrary hidden Markov models.

Function approximation using combined unsupervised and supervised learning.

Science.gov (United States)

Andras, Peter

2014-03-01

Function approximation is one of the core tasks that are solved using neural networks in the context of many engineering problems. However, good approximation results need good sampling of the data space, which usually requires exponentially increasing volume of data as the dimensionality of the data increases. At the same time, often the high-dimensional data is arranged around a much lower dimensional manifold. Here we propose the breaking of the function approximation task for high-dimensional data into two steps: (1) the mapping of the high-dimensional data onto a lower dimensional space corresponding to the manifold on which the data resides and (2) the approximation of the function using the mapped lower dimensional data. We use over-complete self-organizing maps (SOMs) for the mapping through unsupervised learning, and single hidden layer neural networks for the function approximation through supervised learning. We also extend the two-step procedure by considering support vector machines and Bayesian SOMs for the determination of the best parameters for the nonlinear neurons in the hidden layer of the neural networks used for the function approximation. We compare the approximation performance of the proposed neural networks using a set of functions and show that indeed the neural networks using combined unsupervised and supervised learning outperform in most cases the neural networks that learn the function approximation using the original high-dimensional data.

Formulation based on artificial neural network of thermodynamic properties of ozone friendly refrigerant/absorbent couples

International Nuclear Information System (INIS)

Soezen, Adnan; Arcaklioglu, Erol; Oezalp, Mehmet

2005-01-01

This paper presents a new approach based on artificial neural networks (ANNs) to determine the properties of liquid and two phase boiling and condensing of two alternative refrigerant/absorbent couples (methanol/LiBr and methanol/LiCl). These couples do not cause ozone depletion and use in the absorption thermal systems (ATSs). ANNs are able to learn the key information patterns within multidimensional information domain. ANNs operate such as a 'black box' model, requiring no detailed information about the system. On the other hand, they learn the relationship between the input and the output. In order to train the neural network, limited experimental measurements were used as training data and test data. In this study, in input layer, there are temperatures in the range of 298-498 K, pressures (0.1-40 MPa) and concentrations of 2%, 7%, 12% of the couples; specific volume is in output layer. The back-propagation learning algorithm with three different variants, namely scaled conjugate gradient (SCG), Pola-Ribiere conjugate gradient (CGP), and Levenberg-Marquardt (LM), and logistic sigmoid transfer function were used in the network so that the best approach can find. The most suitable algorithm and neuron number in the hidden layer are found as SCG with 8 neurons. For this number level, after the training, it is found that maximum error is less than 3%, average error is about 1% and R 2 value are 99.999%. As seen from the results obtained the thermodynamic equations for each pair by using the weights of network have been obviously predicted within acceptable errors. This paper shows that values predicted with ANN can be used to define the thermodynamic properties instead of approximate and complex analytic equations

A Neural Network Classifier Model for Forecasting Safety Behavior at Workplaces

Directory of Open Access Journals (Sweden)

Fakhradin Ghasemi

2017-07-01

Full Text Available The construction industry is notorious for having an unacceptable rate of fatal accidents. Unsafe behavior has been recognized as the main cause of most accidents occurring at workplaces, particularly construction sites. Having a predictive model of safety behavior can be helpful in preventing construction accidents. The aim of the present study was to build a predictive model of unsafe behavior using the Artificial Neural Network approach. A brief literature review was conducted on factors affecting safe behavior at workplaces and nine factors were selected to be included in the study. Data were gathered using a validated questionnaire from several construction sites. Multilayer perceptron approach was utilized for constructing the desired neural network. Several models with various architectures were tested to find the best one. Sensitivity analysis was conducted to find the most influential factors. The model with one hidden layer containing fourteen hidden neurons demonstrated the best performance (Sum of Squared Errors=6.73. The error rate of the model was approximately 21 percent. The results of sensitivity analysis showed that safety attitude, safety knowledge, supportive environment, and management commitment had the highest effects on safety behavior, while the effects from resource allocation and perceived work pressure were identified to be lower than those of others. The complex nature of human behavior at workplaces and the presence of many influential factors make it difficult to achieve a model with perfect performance.

The Bidirectional Optimization of Carbon Fiber Production by Neural Network with a GA-IPSO Hybrid Algorithm

Directory of Open Access Journals (Sweden)

Jiajia Chen

2013-01-01

Full Text Available A hybrid approach of genetic algorithm (GA and improved particle swarm optimization (IPSO is proposed to construct the radial basis function neural network (RNN for real-time optimizing of the carbon fiber manufacture process. For the three-layer RNN, we adopt the nearest neighbor-clustering algorithm to determine the neurons number of the hidden layer. When the appropriate network structure is fixed, we present the GA-IPSO algorithm to tune the parameters of the network, which means the center and the width of the node in the hidden layer and the weight of output layer. We introduce a penalty factor to adjust the velocity and position of the particles to expedite convergence of the PSO. The GA is used to mutate the particles to escape local optimum. Then we employ this network to develop the bidirectional optimization model: in one direction, we take production parameters as input and properties indices as output; in this case, the model is a carbon fiber product performance prediction system; in the other direction, we take properties indices as input and production parameters as output, and at this situation, the model is a production scheme design tool for novel style carbon fiber. Based on the experimental data, the proposed model is compared to the conventional RBF network and basic PSO method; the research results show its validity and the advantages in dealing with optimization problems.

Fast DCNN based on FWT, intelligent dropout and layer skipping for image retrieval.

Science.gov (United States)

ElAdel, Asma; Zaied, Mourad; Amar, Chokri Ben

2017-11-01

Deep Convolutional Neural Network (DCNN) can be marked as a powerful tool for object and image classification and retrieval. However, the training stage of such networks is highly consuming in terms of storage space and time. Also, the optimization is still a challenging subject. In this paper, we propose a fast DCNN based on Fast Wavelet Transform (FWT), intelligent dropout and layer skipping. The proposed approach led to improve the image retrieval accuracy as well as the searching time. This was possible thanks to three key advantages: First, the rapid way to compute the features using FWT. Second, the proposed intelligent dropout method is based on whether or not a unit is efficiently and not randomly selected. Third, it is possible to classify the image using efficient units of earlier layer(s) and skipping all the subsequent hidden layers directly to the output layer. Our experiments were performed on CIFAR-10 and MNIST datasets and the obtained results are very promising. Copyright © 2017 Elsevier Ltd. All rights reserved.

Optimum Neural Network Architecture for Precipitation Prediction of Myanmar

OpenAIRE

Khaing Win Mar; Thinn Thu Naing

2008-01-01

Nowadays, precipitation prediction is required for proper planning and management of water resources. Prediction with neural network models has received increasing interest in various research and application domains. However, it is difficult to determine the best neural network architecture for prediction since it is not immediately obvious how many input or hidden nodes are used in the model. In this paper, neural network model is used as a forecasting tool. The major aim is to evaluate a s...

Population decoding of motor cortical activity using a generalized linear model with hidden states.

Science.gov (United States)

Lawhern, Vernon; Wu, Wei; Hatsopoulos, Nicholas; Paninski, Liam

2010-06-15

Generalized linear models (GLMs) have been developed for modeling and decoding population neuronal spiking activity in the motor cortex. These models provide reasonable characterizations between neural activity and motor behavior. However, they lack a description of movement-related terms which are not observed directly in these experiments, such as muscular activation, the subject's level of attention, and other internal or external states. Here we propose to include a multi-dimensional hidden state to address these states in a GLM framework where the spike count at each time is described as a function of the hand state (position, velocity, and acceleration), truncated spike history, and the hidden state. The model can be identified by an Expectation-Maximization algorithm. We tested this new method in two datasets where spikes were simultaneously recorded using a multi-electrode array in the primary motor cortex of two monkeys. It was found that this method significantly improves the model-fitting over the classical GLM, for hidden dimensions varying from 1 to 4. This method also provides more accurate decoding of hand state (reducing the mean square error by up to 29% in some cases), while retaining real-time computational efficiency. These improvements on representation and decoding over the classical GLM model suggest that this new approach could contribute as a useful tool to motor cortical decoding and prosthetic applications. Copyright (c) 2010 Elsevier B.V. All rights reserved.

Prediction by artificial neural networks of the physicochemical quality of cane molasses vinegar by time-temperature effect of food to flash evaporator-distiller

Directory of Open Access Journals (Sweden)

Víctor Vásquez V

2010-03-01

Full Text Available It was predicted via Artificial Neural Networks (ANN important physicochemical characteristics of molasses vinegar: pH, density, total acidity, ethanol, total aldehydes and furfural, obtained by flash evaporation operations and flash distillation clarification. Alcoholic and acetic fermented molasses were fed to a flash evaporator at four temperatures (61, 66, 71 and 76 ° C and in three times (25, 35 and 45 min. The prediction was made with two networks: ANN and ANN-A-B, both with good performance. The ANN-A was of the feedforward (FF type with Backpropagation (BP training algorithms and set of Levenberg-Marquardt (LM weights adjustment, topology: 6 inputs (operations data of flash evaporation-distillation, 7 linear outputs (physicochemical characteristics, 9 tangent sigmoidal neurons in 1 hidden layer, 0.5 moment coefficient, 0.01 learning rate, 0.0001 error goal and 20 training stages. The ANN-A showed better performance than a statistical model of first order. The ANN-B also FF, BP and LM algorithms, topology: 2 inputs (data from flash evaporation, 7 linear outputs (physical and chemical characteristics, 84 logarithm sigmoid neurons in 1 hidden layer, 0.5 moment coefficient, 0.01 learning rate, 0.0001 error goal and 300 training stages. The ANN-B showed the same predictive capacity as a statistical model of the first-order with interaction of terms.

Neural Network Based Model of an Industrial Oil-Fired Boiler System ...

African Journals Online (AJOL)

A two-layer feed-forward neural network with Hyperbolic tangent sigmoid ... The neural network model when subjected to test, using the validation input data; ... Proportional Integral Derivative (PID) Controller is used to control the neural ...

Appropriateness of Dropout Layers and Allocation of Their 0.5 Rates across Convolutional Neural Networks for CIFAR-10, EEACL26, and NORB Datasets

Directory of Open Access Journals (Sweden)

Romanuke Vadim V.

2017-12-01

Full Text Available A technique of DropOut for preventing overfitting of convolutional neural networks for image classification is considered in the paper. The goal is to find a rule of rationally allocating DropOut layers of 0.5 rate to maximise performance. To achieve the goal, two common network architectures are used having either 4 or 5 convolutional layers. Benchmarking is fulfilled with CIFAR-10, EEACL26, and NORB datasets. Initially, series of all admissible versions for allocation of DropOut layers are generated. After the performance against the series is evaluated, normalized and averaged, the compromising rule is found. It consists in non-compactly inserting a few DropOut layers before the last convolutional layer. It is likely that the scheme with two or more DropOut layers fits networks of many convolutional layers for image classification problems with a plenty of features. Such a scheme shall also fit simple datasets prone to overfitting. In fact, the rule “prefers” a fewer number of DropOut layers. The exemplary gain of the rule application is roughly between 10 % and 50 %.

Prediction Study on Anti-Slide Control of Railway Vehicle Based on RBF Neural Networks

Science.gov (United States)

Yang, Lijun; Zhang, Jimin

While railway vehicle braking, Anti-slide control system will detect operating status of each wheel-sets e.g. speed difference and deceleration etc. Once the detected value on some wheel-set is over pre-defined threshold, brake effort on such wheel-set will be adjusted automatically to avoid blocking. Such method takes effect on guarantee safety operation of vehicle and avoid wheel-set flatness, however it cannot adapt itself to the rail adhesion variation. While wheel-sets slide, the operating status is chaotic time series with certain law, and can be predicted with the law and experiment data in certain time. The predicted values can be used as the input reference signals of vehicle anti-slide control system, to judge and control the slide status of wheel-sets. In this article, the RBF neural networks is taken to predict wheel-set slide status in multi-step with weight vector adjusted based on online self-adaptive algorithm, and the center & normalizing parameters of active function of the hidden unit of RBF neural networks' hidden layer computed with K-means clustering algorithm. With multi-step prediction simulation, the predicted signal with appropriate precision can be used by anti-slide system to trace actively and adjust wheel-set slide tendency, so as to adapt to wheel-rail adhesion variation and reduce the risk of wheel-set blocking.

Usage of neural network to predict aluminium oxide layer thickness.

Science.gov (United States)

Michal, Peter; Vagaská, Alena; Gombár, Miroslav; Kmec, Ján; Spišák, Emil; Kučerka, Daniel

2015-01-01

This paper shows an influence of chemical composition of used electrolyte, such as amount of sulphuric acid in electrolyte, amount of aluminium cations in electrolyte and amount of oxalic acid in electrolyte, and operating parameters of process of anodic oxidation of aluminium such as the temperature of electrolyte, anodizing time, and voltage applied during anodizing process. The paper shows the influence of those parameters on the resulting thickness of aluminium oxide layer. The impact of these variables is shown by using central composite design of experiment for six factors (amount of sulphuric acid, amount of oxalic acid, amount of aluminium cations, electrolyte temperature, anodizing time, and applied voltage) and by usage of the cubic neural unit with Levenberg-Marquardt algorithm during the results evaluation. The paper also deals with current densities of 1 A · dm(-2) and 3 A · dm(-2) for creating aluminium oxide layer.

Usage of Neural Network to Predict Aluminium Oxide Layer Thickness

Directory of Open Access Journals (Sweden)

Peter Michal

2015-01-01

Full Text Available This paper shows an influence of chemical composition of used electrolyte, such as amount of sulphuric acid in electrolyte, amount of aluminium cations in electrolyte and amount of oxalic acid in electrolyte, and operating parameters of process of anodic oxidation of aluminium such as the temperature of electrolyte, anodizing time, and voltage applied during anodizing process. The paper shows the influence of those parameters on the resulting thickness of aluminium oxide layer. The impact of these variables is shown by using central composite design of experiment for six factors (amount of sulphuric acid, amount of oxalic acid, amount of aluminium cations, electrolyte temperature, anodizing time, and applied voltage and by usage of the cubic neural unit with Levenberg-Marquardt algorithm during the results evaluation. The paper also deals with current densities of 1 A·dm−2 and 3 A·dm−2 for creating aluminium oxide layer.

Exploring the Combination of Dempster-Shafer Theory and Neural Network for Predicting Trust and Distrust

Directory of Open Access Journals (Sweden)

Xin Wang

2016-01-01

Full Text Available In social media, trust and distrust among users are important factors in helping users make decisions, dissect information, and receive recommendations. However, the sparsity and imbalance of social relations bring great difficulties and challenges in predicting trust and distrust. Meanwhile, there are numerous inducing factors to determine trust and distrust relations. The relationship among inducing factors may be dependency, independence, and conflicting. Dempster-Shafer theory and neural network are effective and efficient strategies to deal with these difficulties and challenges. In this paper, we study trust and distrust prediction based on the combination of Dempster-Shafer theory and neural network. We firstly analyze the inducing factors about trust and distrust, namely, homophily, status theory, and emotion tendency. Then, we quantify inducing factors of trust and distrust, take these features as evidences, and construct evidence prototype as input nodes of multilayer neural network. Finally, we propose a framework of predicting trust and distrust which uses multilayer neural network to model the implementing process of Dempster-Shafer theory in different hidden layers, aiming to overcome the disadvantage of Dempster-Shafer theory without optimization method. Experimental results on a real-world dataset demonstrate the effectiveness of the proposed framework.

Meta-modeling of the pesticide fate model MACRO for groundwater exposure assessments using artificial neural networks

Science.gov (United States)

Stenemo, Fredrik; Lindahl, Anna M. L.; Gärdenäs, Annemieke; Jarvis, Nicholas

2007-08-01

Several simple index methods that use easily accessible data have been developed and included in decision-support systems to estimate pesticide leaching across larger areas. However, these methods often lack important process descriptions (e.g. macropore flow), which brings into question their reliability. Descriptions of macropore flow have been included in simulation models, but these are too complex and demanding for spatial applications. To resolve this dilemma, a neural network simulation meta-model of the dual-permeability macropore flow model MACRO was created for pesticide groundwater exposure assessment. The model was parameterized using pedotransfer functions that require as input the clay and sand content of the topsoil and subsoil, and the topsoil organic carbon content. The meta-model also requires the topsoil pesticide half-life and the soil organic carbon sorption coefficient as input. A fully connected feed-forward multilayer perceptron classification network with two hidden layers, linked to fully connected feed-forward multilayer perceptron neural networks with one hidden layer, trained on sub-sets of the target variable, was shown to be a suitable meta-model for the intended purpose. A Fourier amplitude sensitivity test showed that the model output (the 80th percentile average yearly pesticide concentration at 1 m depth for a 20 year simulation period) was sensitive to all input parameters. The two input parameters related to pesticide characteristics (i.e. soil organic carbon sorption coefficient and topsoil pesticide half-life) were the most influential, but texture in the topsoil was also quite important since it was assumed to control the mass exchange coefficient that regulates the strength of macropore flow. This is in contrast to models based on the advection-dispersion equation where soil texture is relatively unimportant. The use of the meta-model is exemplified with a case-study where the spatial variability of pesticide leaching is

Klasifikasi Varietas Cabai Berdasarkan Morfologi Daun Menggunakan Backpropagation Neural Network

Directory of Open Access Journals (Sweden)

Kharis Syaban

2016-07-01

Full Text Available Compared with other methods of classifiers such as cellular and molecular biological methods, using the image of the leaves become the first choice in the classification of plants. The leaves can be characterized by shape, color, and texture; The leaves can have a color that varies depending on the season and geographical location. In addition, the same plant species also can have different leaf shapes. In this study, the morphological features of leaves used to identify varieties of pepper plants. The method used to perform feature extraction is a moment invariant and basic geometric features. For the process of recognition based on the features that have been extracted, used neural network methods with backpropagation learning algorithm. From the neural-network training, the best accuracy in classifying varieties of chili with minimum error 0.001 by providing learning rate 0.1, momentum of 0.7, and 15 neurons in the hidden layer foreach of various feature. To conduct cross-validation testing with k-fold tehcnique, obtained classification accuracy to be range of 80.75%±0.09% with k=4.

Compressing the hidden variable space of a qubit

International Nuclear Information System (INIS)

Montina, Alberto

2011-01-01

In previously exhibited hidden variable models of quantum state preparation and measurement, the number of continuous hidden variables describing the actual state of single realizations is never smaller than the quantum state manifold dimension. We introduce a simple model for a qubit whose hidden variable space is one-dimensional, i.e., smaller than the two-dimensional Bloch sphere. The hidden variable probability distributions associated with quantum states satisfy reasonable criteria of regularity. Possible generalizations of this shrinking to an N-dimensional Hilbert space are discussed.

Penentuan Error Dalam Peramalan Jumlah Korban Demam Berdarah Dengue Menggunakan Metode Neural Network (Kasus : Rumah Sakit Charitas Palembang

Directory of Open Access Journals (Sweden)

Maria Bellaniar Ismiati

2016-11-01

Full Text Available Dengue Hemorrhagic Fever (DHF is a type of disease that was ranked first in ASEAN and ranked second in the world. The number of victims of dengue in RS Charitas Palembang tend to increase in certain months and erratic every month. In addition, dengue casualty data is not used as an evaluation to reduce the number of victims. It became the basis for forecasting the number of victims of dengue in the next year. Research to predict the number of victims of dengue have been done with various techniques of artificial intelligence. Research conducted now use data RS Charitas Palembang patterned time series over the last 10 years by using Neural Network. The results obtained are patterns victim DBD significant start in December and then reach the peak in January, accompanied by figures forecast in each month of the following year. Furthermore, the calculation error using Neural Network obtained using the input layer 12, hidden neuron 28, and the output layer 1 and the error obtained 12.59%.

Aspects of artificial neural networks - with applications in high energy physics

International Nuclear Information System (INIS)

Roegnvaldsson, T.S.

1994-02-01

Different aspects of artificial neural networks are studied and discussed. They are demonstrated to be powerful general purpose algorithms, applicable to many different problem areas like pattern recognition, function fitting and prediction. Multi-layer perceptron (MPL) models are shown to out perform previous standard approaches on both off-line and on-line analysis tasks in high energy physics, like quark flavour tagging and mass reconstruction, as well as being powerful tools for prediction tasks. It is also demonstrated how a self-organizing network can be employed to extract information from data, for instance to track down origins of unexpected model discrepancies. Furthermore, it is proved that the MPL is more efficient than the learning vector quantization technique on classification problems, by producing smoother discrimination surfaces, and that an MPL network should be trained with a noisy updating schedule if the Hessian is ill-conditioned - A result that is especially important for MPL network with more than just one hidden layer. 81 refs, 6 figs

Application of Neural Networks for classification of Patau, Edwards, Down, Turner and Klinefelter Syndrome based on first trimester maternal serum screening data, ultrasonographic findings and patient demographics.

Science.gov (United States)

Catic, Aida; Gurbeta, Lejla; Kurtovic-Kozaric, Amina; Mehmedbasic, Senad; Badnjevic, Almir

2018-02-13

The usage of Artificial Neural Networks (ANNs) for genome-enabled classifications and establishing genome-phenotype correlations have been investigated more extensively over the past few years. The reason for this is that ANNs are good approximates of complex functions, so classification can be performed without the need for explicitly defined input-output model. This engineering tool can be applied for optimization of existing methods for disease/syndrome classification. Cytogenetic and molecular analyses are the most frequent tests used in prenatal diagnostic for the early detection of Turner, Klinefelter, Patau, Edwards and Down syndrome. These procedures can be lengthy, repetitive; and often employ invasive techniques so a robust automated method for classifying and reporting prenatal diagnostics would greatly help the clinicians with their routine work. The database consisted of data collected from 2500 pregnant woman that came to the Institute of Gynecology, Infertility and Perinatology "Mehmedbasic" for routine antenatal care between January 2000 and December 2016. During first trimester all women were subject to screening test where values of maternal serum pregnancy-associated plasma protein A (PAPP-A) and free beta human chorionic gonadotropin (β-hCG) were measured. Also, fetal nuchal translucency thickness and the presence or absence of the nasal bone was observed using ultrasound. The architectures of linear feedforward and feedback neural networks were investigated for various training data distributions and number of neurons in hidden layer. Feedback neural network architecture out performed feedforward neural network architecture in predictive ability for all five aneuploidy prenatal syndrome classes. Feedforward neural network with 15 neurons in hidden layer achieved classification sensitivity of 92.00%. Classification sensitivity of feedback (Elman's) neural network was 99.00%. Average accuracy of feedforward neural network was 89.6% and for

Cox-nnet: An artificial neural network method for prognosis prediction of high-throughput omics data.

Science.gov (United States)

Ching, Travers; Zhu, Xun; Garmire, Lana X

2018-04-01

Artificial neural networks (ANN) are computing architectures with many interconnections of simple neural-inspired computing elements, and have been applied to biomedical fields such as imaging analysis and diagnosis. We have developed a new ANN framework called Cox-nnet to predict patient prognosis from high throughput transcriptomics data. In 10 TCGA RNA-Seq data sets, Cox-nnet achieves the same or better predictive accuracy compared to other methods, including Cox-proportional hazards regression (with LASSO, ridge, and mimimax concave penalty), Random Forests Survival and CoxBoost. Cox-nnet also reveals richer biological information, at both the pathway and gene levels. The outputs from the hidden layer node provide an alternative approach for survival-sensitive dimension reduction. In summary, we have developed a new method for accurate and efficient prognosis prediction on high throughput data, with functional biological insights. The source code is freely available at https://github.com/lanagarmire/cox-nnet.

Dynamics of neural cryptography.

Science.gov (United States)

Ruttor, Andreas; Kinzel, Wolfgang; Kanter, Ido

2007-05-01

Synchronization of neural networks has been used for public channel protocols in cryptography. In the case of tree parity machines the dynamics of both bidirectional synchronization and unidirectional learning is driven by attractive and repulsive stochastic forces. Thus it can be described well by a random walk model for the overlap between participating neural networks. For that purpose transition probabilities and scaling laws for the step sizes are derived analytically. Both these calculations as well as numerical simulations show that bidirectional interaction leads to full synchronization on average. In contrast, successful learning is only possible by means of fluctuations. Consequently, synchronization is much faster than learning, which is essential for the security of the neural key-exchange protocol. However, this qualitative difference between bidirectional and unidirectional interaction vanishes if tree parity machines with more than three hidden units are used, so that those neural networks are not suitable for neural cryptography. In addition, the effective number of keys which can be generated by the neural key-exchange protocol is calculated using the entropy of the weight distribution. As this quantity increases exponentially with the system size, brute-force attacks on neural cryptography can easily be made unfeasible.

Dynamics of neural cryptography

International Nuclear Information System (INIS)

Ruttor, Andreas; Kinzel, Wolfgang; Kanter, Ido

2007-01-01

Synchronization of neural networks has been used for public channel protocols in cryptography. In the case of tree parity machines the dynamics of both bidirectional synchronization and unidirectional learning is driven by attractive and repulsive stochastic forces. Thus it can be described well by a random walk model for the overlap between participating neural networks. For that purpose transition probabilities and scaling laws for the step sizes are derived analytically. Both these calculations as well as numerical simulations show that bidirectional interaction leads to full synchronization on average. In contrast, successful learning is only possible by means of fluctuations. Consequently, synchronization is much faster than learning, which is essential for the security of the neural key-exchange protocol. However, this qualitative difference between bidirectional and unidirectional interaction vanishes if tree parity machines with more than three hidden units are used, so that those neural networks are not suitable for neural cryptography. In addition, the effective number of keys which can be generated by the neural key-exchange protocol is calculated using the entropy of the weight distribution. As this quantity increases exponentially with the system size, brute-force attacks on neural cryptography can easily be made unfeasible

Dynamics of neural cryptography

Science.gov (United States)

Ruttor, Andreas; Kinzel, Wolfgang; Kanter, Ido

2007-05-01

Synchronization of neural networks has been used for public channel protocols in cryptography. In the case of tree parity machines the dynamics of both bidirectional synchronization and unidirectional learning is driven by attractive and repulsive stochastic forces. Thus it can be described well by a random walk model for the overlap between participating neural networks. For that purpose transition probabilities and scaling laws for the step sizes are derived analytically. Both these calculations as well as numerical simulations show that bidirectional interaction leads to full synchronization on average. In contrast, successful learning is only possible by means of fluctuations. Consequently, synchronization is much faster than learning, which is essential for the security of the neural key-exchange protocol. However, this qualitative difference between bidirectional and unidirectional interaction vanishes if tree parity machines with more than three hidden units are used, so that those neural networks are not suitable for neural cryptography. In addition, the effective number of keys which can be generated by the neural key-exchange protocol is calculated using the entropy of the weight distribution. As this quantity increases exponentially with the system size, brute-force attacks on neural cryptography can easily be made unfeasible.

A classification of hidden-variable properties

International Nuclear Information System (INIS)

Brandenburger, Adam; Yanofsky, Noson

2008-01-01

Hidden variables are extra components added to try to banish counterintuitive features of quantum mechanics. We start with a quantum-mechanical model and describe various properties that can be asked of a hidden-variable model. We present six such properties and a Venn diagram of how they are related. With two existence theorems and three no-go theorems (EPR, Bell and Kochen-Specker), we show which properties of empirically equivalent hidden-variable models are possible and which are not. Formally, our treatment relies only on classical probability models, and physical phenomena are used only to motivate which models to choose

A comparative study of two neural networks for document retrieval

International Nuclear Information System (INIS)

Hui, S.C.; Goh, A.

1997-01-01

In recent years there has been specific interest in adopting advanced computer techniques in the field of document retrieval. This interest is generated by the fact that classical methods such as the Boolean search, the vector space model or even probabilistic retrieval cannot handle the increasing demands of end-users in satisfying their needs. The most recent attempt is the application of the neural network paradigm as a means of providing end-users with a more powerful retrieval mechanism. Neural networks are not only good pattern matchers but also highly versatile and adaptable. In this paper, we demonstrate how to apply two neural networks, namely Adaptive Resonance Theory and Fuzzy Kohonen Neural Network, for document retrieval. In addition, a comparison of these two neural networks based on performance is also given

Deep neural nets as a method for quantitative structure-activity relationships.

Science.gov (United States)

Ma, Junshui; Sheridan, Robert P; Liaw, Andy; Dahl, George E; Svetnik, Vladimir

2015-02-23

Neural networks were widely used for quantitative structure-activity relationships (QSAR) in the 1990s. Because of various practical issues (e.g., slow on large problems, difficult to train, prone to overfitting, etc.), they were superseded by more robust methods like support vector machine (SVM) and random forest (RF), which arose in the early 2000s. The last 10 years has witnessed a revival of neural networks in the machine learning community thanks to new methods for preventing overfitting, more efficient training algorithms, and advancements in computer hardware. In particular, deep neural nets (DNNs), i.e. neural nets with more than one hidden layer, have found great successes in many applications, such as computer vision and natural language processing. Here we show that DNNs can routinely make better prospective predictions than RF on a set of large diverse QSAR data sets that are taken from Merck's drug discovery effort. The number of adjustable parameters needed for DNNs is fairly large, but our results show that it is not necessary to optimize them for individual data sets, and a single set of recommended parameters can achieve better performance than RF for most of the data sets we studied. The usefulness of the parameters is demonstrated on additional data sets not used in the calibration. Although training DNNs is still computationally intensive, using graphical processing units (GPUs) can make this issue manageable.

Application of structured support vector machine backpropagation to a convolutional neural network for human pose estimation.

Science.gov (United States)

Witoonchart, Peerajak; Chongstitvatana, Prabhas

2017-08-01

In this study, for the first time, we show how to formulate a structured support vector machine (SSVM) as two layers in a convolutional neural network, where the top layer is a loss augmented inference layer and the bottom layer is the normal convolutional layer. We show that a deformable part model can be learned with the proposed structured SVM neural network by backpropagating the error of the deformable part model to the convolutional neural network. The forward propagation calculates the loss augmented inference and the backpropagation calculates the gradient from the loss augmented inference layer to the convolutional layer. Thus, we obtain a new type of convolutional neural network called an Structured SVM convolutional neural network, which we applied to the human pose estimation problem. This new neural network can be used as the final layers in deep learning. Our method jointly learns the structural model parameters and the appearance model parameters. We implemented our method as a new layer in the existing Caffe library. Copyright © 2017 Elsevier Ltd. All rights reserved.

A simple approach for the sonochemical loading of Au, Ag and Pd nanoparticle on functionalized MWCNT and subsequent dispersion studies for removal of organic dyes: Artificial neural network and response surface methodology studies.

Science.gov (United States)

Moghaddari, Mitra; Yousefi, Fakhri; Ghaedi, Mehrorang; Dashtian, Kheibar

2018-04-01

In this study, the artificial neural network (ANN) and response surface methodology (RSM) based on central composite design (CCD) were applied for modeling and optimization of the simultaneous ultrasound-assisted removal of quinoline yellow (QY) and eosin B (EB). The MWCNT-NH 2 and its composites were prepared by sonochemistry method and characterized by scanning electron microscopy (SEM), X-ray diffraction (XRD) and energy dispersive spectroscopy (EDS) analysis's. Initial dyes concentrations, adsorbent mass, sonication time and pH contribution on QY and EB removal percentage were investigated by CCD and replication of experiments at conditions suggested by model has results which statistically are close to experimented data. The ultrasound irradiation is associated with raising mass transfer of process so that small amount of the adsorbent (0.025 g) is able to remove high percentage (88.00% and 91.00%) of QY and EB, respectively in short time (6.0 min) at pH = 6. Analysis of experimental data by conventional models is good indication of Langmuir efficiency for fitting and explanation of experimented data. The ANN based on the Levenberg-Marquardt algorithm (LMA) combined of linear transfer function at output layer and tangent sigmoid transfer function at hidden layer with 20 hidden neurons supply best operation conditions for good prediction of adsorption data. Accurate and efficient artificial neural network was obtained by changing the number of neurons in the hidden layer, while data was divided into training, test and validation sets which contained 70, 15 and 15% of data points respectively. The Average absolute deviation (AAD)% of a collection of 128 data points for MWCNT-NH 2 and composites is 0.58%.for EB and 0.55 for YQ. Copyright © 2017 Elsevier B.V. All rights reserved.

Sociology of Hidden Curriculum

Directory of Open Access Journals (Sweden)

Alireza Moradi

2017-06-01

Full Text Available This paper reviews the concept of hidden curriculum in the sociological theories and wants to explain sociological aspects of formation of hidden curriculum. The main question concentrates on the theoretical approaches in which hidden curriculum is explained sociologically.For this purpose it was applied qualitative research methodology. The relevant data include various sociological concepts and theories of hidden curriculum collected by the documentary method. The study showed a set of rules, procedures, relationships and social structure of education have decisive role in the formation of hidden curriculum. A hidden curriculum reinforces by existed inequalities among learners (based on their social classes or statues. There is, in fact, a balance between the learner's "knowledge receptions" with their "inequality proportion".The hidden curriculum studies from different major sociological theories such as Functionalism, Marxism and critical theory, Symbolic internationalism and Feminism. According to the functionalist perspective a hidden curriculum has a social function because it transmits social values. Marxists and critical thinkers correlate between hidden curriculum and the totality of social structure. They depicts that curriculum prepares learners for the exploitation in the work markets. Symbolic internationalism rejects absolute hegemony of hidden curriculum on education and looks to the socialization as a result of interaction between learner and instructor. Feminism theory also considers hidden curriculum as a vehicle which legitimates gender stereotypes.

Estimation of Leakage Ratio Using Principal Component Analysis and Artificial Neural Network in Water Distribution Systems

Directory of Open Access Journals (Sweden)

Dongwoo Jang

2018-03-01

Full Text Available Leaks in a water distribution network (WDS constitute losses of water supply caused by pipeline failure, operational loss, and physical factors. This has raised the need for studies on the factors affecting the leakage ratio and estimation of leakage volume in a water supply system. In this study, principal component analysis (PCA and artificial neural network (ANN were used to estimate the volume of water leakage in a WDS. For the study, six main effective parameters were selected and standardized data obtained through the Z-score method. The PCA-ANN model was devised and the leakage ratio was estimated. An accuracy assessment was performed to compare the measured leakage ratio to that of the simulated model. The results showed that the PCA-ANN method was more accurate for estimating the leakage ratio than a single ANN simulation. In addition, the estimation results differed according to the number of neurons in the ANN model’s hidden layers. In this study, an ANN with multiple hidden layers was found to be the best method for estimating the leakage ratio with 12–12 neurons. This suggested approaches to improve the accuracy of leakage ratio estimation, as well as a scientific approach toward the sustainable management of water distribution systems.

Evaluation of extra virgin olive oil stability by artificial neural network.

Science.gov (United States)

Silva, Simone Faria; Anjos, Carlos Alberto Rodrigues; Cavalcanti, Rodrigo Nunes; Celeghini, Renata Maria dos Santos

2015-07-15

The stability of extra virgin olive oil in polyethylene terephthalate bottles and tinplate cans stored for 6 months under dark and light conditions was evaluated. The following analyses were carried out: free fatty acids, peroxide value, specific extinction at 232 and 270 nm, chlorophyll, L(∗)C(∗)h color, total phenolic compounds, tocopherols and squalene. The physicochemical changes were evaluated by artificial neural network (ANN) modeling with respect to light exposure conditions and packaging material. The optimized ANN structure consists of 11 input neurons, 18 hidden neurons and 5 output neurons using hyperbolic tangent and softmax activation functions in hidden and output layers, respectively. The five output neurons correspond to five possible classifications according to packaging material (PET amber, PET transparent and tinplate can) and light exposure (dark and light storage). The predicted physicochemical changes agreed very well with the experimental data showing high classification accuracy for test (>90%) and training set (>85). Sensitivity analysis showed that free fatty acid content, peroxide value, L(∗)Cab(∗)hab(∗) color parameters, tocopherol and chlorophyll contents were the physicochemical attributes with the most discriminative power. Copyright © 2015 Elsevier Ltd. All rights reserved.

Foreground removal from CMB temperature maps using an MLP neural network

Science.gov (United States)

Nørgaard-Nielsen, H. U.; Jørgensen, H. E.

2008-12-01

One of the main obstacles for extracting the Cosmic Microwave Background (CMB) signal from observations in the mm-submm range is the foreground contamination by emission from Galactic components: mainly synchrotron, free-free and thermal dust emission. Due to the statistical nature of the intrinsic CMB signal it is essential to minimize the systematic errors in the CMB temperature determinations. Following the available knowledge of the spectral behavior of the Galactic foregrounds simple power law-like spectra have been assumed. The feasibility of using a simple neural network for extracting the CMB temperature signal from the combined signal CMB and the foregrounds has been investigated. As a specific example, we have analysed simulated data, as expected from the ESA Planck CMB mission. A simple multilayer perceptron neural network with 2 hidden layers can provide temperature estimates over more than 80 per cent of the sky that are to a high degree uncorrelated with the foreground signals. A single network will be able to cover the dynamic range of the Planck noise level over the entire sky.

Detection of Pistachio Aflatoxin Using Raman Spectroscopy and Artificial Neural Networks

Directory of Open Access Journals (Sweden)

R Mohammadigol

2015-03-01

Full Text Available Pistachio contamination to aflatoxin has been known as a serious problem for pistachio exportation. With regards to the increasing demand for Raman spectroscopy to detect and classify different materials and also the current experimental and technical problems for measuring toxin (such as being expensive and time-consuming, the main objective of this study was to detect aflatoxin contamination in pistachio by using Raman spectroscopy technique and artificial neural networks. Three sets of samples were prepared: non-contaminated (healthy and contaminated samples with 20 and 100 ppb of the total aflatoxins (B1+B2+G1+G2. After spectral acquisition, considering to the results, spectral data were normalized and then principal components (PCs were extracted to reduce the data dimensions. For classification of the samples spectra, an artificial neural network was used with a feed forward back propagation algorithm for 4 inputs and 3 neurons in hidden layer. Mean overall accuracy was achieved to be 98 percent; therefore, non-liner Raman spectra data modeling by ANN for samples classification was successful.

Learning of N-layers neural network

Directory of Open Access Journals (Sweden)

Vladimír Konečný

2005-01-01

Full Text Available In the last decade we can observe increasing number of applications based on the Artificial Intelligence that are designed to solve problems from different areas of human activity. The reason why there is so much interest in these technologies is that the classical way of solutions does not exist or these technologies are not suitable because of their robustness. They are often used in applications like Business Intelligence that enable to obtain useful information for high-quality decision-making and to increase competitive advantage.One of the most widespread tools for the Artificial Intelligence are the artificial neural networks. Their high advantage is relative simplicity and the possibility of self-learning based on set of pattern situations.For the learning phase is the most commonly used algorithm back-propagation error (BPE. The base of BPE is the method minima of error function representing the sum of squared errors on outputs of neural net, for all patterns of the learning set. However, while performing BPE and in the first usage, we can find out that it is necessary to complete the handling of the learning factor by suitable method. The stability of the learning process and the rate of convergence depend on the selected method. In the article there are derived two functions: one function for the learning process management by the relative great error function value and the second function when the value of error function approximates to global minimum.The aim of the article is to introduce the BPE algorithm in compact matrix form for multilayer neural networks, the derivation of the learning factor handling method and the presentation of the results.

A Novel Approach to ECG Classification Based upon Two-Layered HMMs in Body Sensor Networks

Science.gov (United States)

Liang, Wei; Zhang, Yinlong; Tan, Jindong; Li, Yang

2014-01-01

This paper presents a novel approach to ECG signal filtering and classification. Unlike the traditional techniques which aim at collecting and processing the ECG signals with the patient being still, lying in bed in hospitals, our proposed algorithm is intentionally designed for monitoring and classifying the patient's ECG signals in the free-living environment. The patients are equipped with wearable ambulatory devices the whole day, which facilitates the real-time heart attack detection. In ECG preprocessing, an integral-coefficient-band-stop (ICBS) filter is applied, which omits time-consuming floating-point computations. In addition, two-layered Hidden Markov Models (HMMs) are applied to achieve ECG feature extraction and classification. The periodic ECG waveforms are segmented into ISO intervals, P subwave, QRS complex and T subwave respectively in the first HMM layer where expert-annotation assisted Baum-Welch algorithm is utilized in HMM modeling. Then the corresponding interval features are selected and applied to categorize the ECG into normal type or abnormal type (PVC, APC) in the second HMM layer. For verifying the effectiveness of our algorithm on abnormal signal detection, we have developed an ECG body sensor network (BSN) platform, whereby real-time ECG signals are collected, transmitted, displayed and the corresponding classification outcomes are deduced and shown on the BSN screen. PMID:24681668

Accurate estimation of CO2 adsorption on activated carbon with multi-layer feed-forward neural network (MLFNN algorithm

Directory of Open Access Journals (Sweden)

Alireza Rostami

2018-03-01

Full Text Available Global warming due to greenhouse effect has been considered as a serious problem for many years around the world. Among the different gases which cause greenhouse gas effect, carbon dioxide is of great difficulty by entering into the surrounding atmosphere. So CO2 capturing and separation especially by adsorption is one of the most interesting approaches because of the low equipment cost, ease of operation, simplicity of design, and low energy consumption.In this study, experimental results are presented for the adsorption equilibria of carbon dioxide on activated carbon. The adsorption equilibrium data for carbon dioxide were predicted with two commonly used isotherm models in order to compare with multi-layer feed-forward neural network (MLFNN algorithm for a wide range of partial pressure. As a result, the ANN-based algorithm shows much better efficiency and accuracy than the Sips and Langmuir isotherms. In addition, the applicability of the Sips and Langmuir models are limited to isothermal conditions, even though the ANN-based algorithm is not restricted to the constant temperature condition. Consequently, it is proved that MLFNN algorithm is a promising model for calculation of CO2 adsorption density on activated carbon. Keywords: Global warming, CO2 adsorption, Activated carbon, Multi-layer feed-forward neural network algorithm, Statistical quality measures

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

Science.gov (United States)

Chen, Fangyue; Chen, Guanrong Ron; He, Guolong; Xu, Xiubin; He, Qinbin

2009-10-01

Universal perceptron (UP), a generalization of Rosenblatt's perceptron, is considered in this paper, which is capable of implementing all Boolean functions (BFs). In the classification of BFs, there are: 1) linearly separable Boolean function (LSBF) class, 2) parity Boolean function (PBF) class, and 3) non-LSBF and non-PBF class. To implement these functions, UP takes different kinds of simple topological structures in which each contains at most one hidden layer along with the smallest possible number of hidden neurons. Inspired by the concept of DNA sequences in biological systems, a novel learning algorithm named DNA-like learning is developed, which is able to quickly train a network with any prescribed BF. The focus is on performing LSBF and PBF by a single-layer perceptron (SLP) with the new algorithm. Two criteria for LSBF and PBF are proposed, respectively, and a new measure for a BF, named nonlinearly separable degree (NLSD), is introduced. In the sense of this measure, the PBF is the most complex one. The new algorithm has many advantages including, in particular, fast running speed, good robustness, and no need of considering the convergence property. For example, the number of iterations and computations in implementing the basic 2-bit logic operations such as AND, OR, and XOR by using the new algorithm is far smaller than the ones needed by using other existing algorithms such as error-correction (EC) and backpropagation (BP) algorithms. Moreover, the synaptic weights and threshold values derived from UP can be directly used in designing of the template of cellular neural networks (CNNs), which has been considered as a new spatial-temporal sensory computing paradigm.

A microsensor array for quantification of lubricant contaminants using a back propagation artificial neural network

International Nuclear Information System (INIS)

Zhu, Xiaoliang; Du, Li; Zhe, Jiang; Liu, Bendong

2016-01-01

We present a method based on an electrochemical sensor array and a back propagation artificial neural network for detection and quantification of four properties of lubrication oil, namely water (0, 500 ppm, 1000 ppm), total acid number (TAN) (13.1, 13.7, 14.4, 15.6 mg KOH g −1 ), soot (0, 1%, 2%, 3%) and sulfur content (1.3%, 1.37%, 1.44%, 1.51%). The sensor array, consisting of four micromachined electrochemical sensors, detects the four properties with overlapping sensitivities. A total set of 36 oil samples containing mixtures of water, soot, and sulfuric acid with different concentrations were prepared for testing. The sensor array’s responses were then divided to three sets: training sets (80% data), validation sets (10%) and testing sets (10%). Several back propagation artificial neural network architectures were trained with the training and validation sets; one architecture with four input neurons, 50 and 5 neurons in the first and second hidden layer, and four neurons in the output layer was selected. The selected neural network was then tested using the four sets of testing data (10%). Test results demonstrated that the developed artificial neural network is able to quantitatively determine the four lubrication properties (water, TAN, soot, and sulfur content) with a maximum prediction error of 18.8%, 6.0%, 6.7%, and 5.4%, respectively, indicting a good match between the target and predicted values. With the developed network, the sensor array could be potentially used for online lubricant oil condition monitoring. (paper)

Hidden measurements, hidden variables and the volume representation of transition probabilities

OpenAIRE

Oliynyk, Todd A.

2005-01-01

We construct, for any finite dimension $n$, a new hidden measurement model for quantum mechanics based on representing quantum transition probabilities by the volume of regions in projective Hilbert space. For $n=2$ our model is equivalent to the Aerts sphere model and serves as a generalization of it for dimensions $n \\geq 3$. We also show how to construct a hidden variables scheme based on hidden measurements and we discuss how joint distributions arise in our hidden variables scheme and th...

Flood forecasting for the upper reach of the Red River Basin, North ...

African Journals Online (AJOL)

This is a multilayer feed forward neural network with a back- ... 2 with the input layer, one hidden layer, andthe output layer with only one node . Theactivation ..... Thepaper is aresult of theresearch project entitled FLOod Control. Decision ...

Genetic algorithm for neural networks optimization

Science.gov (United States)

Setyawati, Bina R.; Creese, Robert C.; Sahirman, Sidharta

2004-11-01

This paper examines the forecasting performance of multi-layer feed forward neural networks in modeling a particular foreign exchange rates, i.e. Japanese Yen/US Dollar. The effects of two learning methods, Back Propagation and Genetic Algorithm, in which the neural network topology and other parameters fixed, were investigated. The early results indicate that the application of this hybrid system seems to be well suited for the forecasting of foreign exchange rates. The Neural Networks and Genetic Algorithm were programmed using MATLAB«.

Artificial neural network combined with principal component analysis for resolution of complex pharmaceutical formulations.

Science.gov (United States)

Ioele, Giuseppina; De Luca, Michele; Dinç, Erdal; Oliverio, Filomena; Ragno, Gaetano

2011-01-01

A chemometric approach based on the combined use of the principal component analysis (PCA) and artificial neural network (ANN) was developed for the multicomponent determination of caffeine (CAF), mepyramine (MEP), phenylpropanolamine (PPA) and pheniramine (PNA) in their pharmaceutical preparations without any chemical separation. The predictive ability of the ANN method was compared with the classical linear regression method Partial Least Squares 2 (PLS2). The UV spectral data between 220 and 300 nm of a training set of sixteen quaternary mixtures were processed by PCA to reduce the dimensions of input data and eliminate the noise coming from instrumentation. Several spectral ranges and different numbers of principal components (PCs) were tested to find the PCA-ANN and PLS2 models reaching the best determination results. A two layer ANN, using the first four PCs, was used with log-sigmoid transfer function in first hidden layer and linear transfer function in output layer. Standard error of prediction (SEP) was adopted to assess the predictive accuracy of the models when subjected to external validation. PCA-ANN showed better prediction ability in the determination of PPA and PNA in synthetic samples with added excipients and pharmaceutical formulations. Since both components are characterized by low absorptivity, the better performance of PCA-ANN was ascribed to the ability in considering all non-linear information from noise or interfering excipients.

Hidden Markov models and other machine learning approaches in computational molecular biology

Energy Technology Data Exchange (ETDEWEB)

Baldi, P. [California Inst. of Tech., Pasadena, CA (United States)

1995-12-31

This tutorial was one of eight tutorials selected to be presented at the Third International Conference on Intelligent Systems for Molecular Biology which was held in the United Kingdom from July 16 to 19, 1995. Computational tools are increasingly needed to process the massive amounts of data, to organize and classify sequences, to detect weak similarities, to separate coding from non-coding regions, and reconstruct the underlying evolutionary history. The fundamental problem in machine learning is the same as in scientific reasoning in general, as well as statistical modeling: to come up with a good model for the data. In this tutorial four classes of models are reviewed. They are: Hidden Markov models; artificial Neural Networks; Belief Networks; and Stochastic Grammars. When dealing with DNA and protein primary sequences, Hidden Markov models are one of the most flexible and powerful alignments and data base searches. In this tutorial, attention is focused on the theory of Hidden Markov Models, and how to apply them to problems in molecular biology.

Fitting Hidden Markov Models to Psychological Data

Directory of Open Access Journals (Sweden)

Ingmar Visser

2002-01-01

Full Text Available Markov models have been used extensively in psychology of learning. Applications of hidden Markov models are rare however. This is partially due to the fact that comprehensive statistics for model selection and model assessment are lacking in the psychological literature. We present model selection and model assessment statistics that are particularly useful in applying hidden Markov models in psychology. These statistics are presented and evaluated by simulation studies for a toy example. We compare AIC, BIC and related criteria and introduce a prediction error measure for assessing goodness-of-fit. In a simulation study, two methods of fitting equality constraints are compared. In two illustrative examples with experimental data we apply selection criteria, fit models with constraints and assess goodness-of-fit. First, data from a concept identification task is analyzed. Hidden Markov models provide a flexible approach to analyzing such data when compared to other modeling methods. Second, a novel application of hidden Markov models in implicit learning is presented. Hidden Markov models are used in this context to quantify knowledge that subjects express in an implicit learning task. This method of analyzing implicit learning data provides a comprehensive approach for addressing important theoretical issues in the field.

Assessing artificial neural network performance in estimating the layer properties of pavements

Directory of Open Access Journals (Sweden)

Gloria Inés Beltran

2014-05-01

Full Text Available A major concern in assessing the structural condition of existing flexible pavements is the estimation of the mechanical properties of constituent layers, which is useful for the design and decision-making process in road management systems. This parameter identification problem is truly complex due to the large number of variables involved in pavement behavior. To this end, non-conventional adaptive or approximate solutions via Artificial Neural Networks – ANNs – are considered to properly map pavement response field measurements. Previous investigations have demonstrated the exceptional ability of ANNs in layer moduli estimation from non-destructive deflection tests, but most of the reported cases were developed using synthetic deflection data or hypothetical pavement systems. This paper presents further attempts to back-calculate layer moduli via ANN modeling, using a database gathered from field tests performed on three- and four-layer pavement systems. Traditional layer structuring and pavements with a stabilized subbase were considered. A three-stage methodology is developed in this study to design and validate an “optimum” ANN-based model, i.e., the best architecture possible along with adequate learning rules. An assessment of the resulting ANN model demonstrates its forecasting capabilities and efficiency in solving a complex parameter identification problem concerning pavements.

Optimization of recurrent neural networks for time series modeling

DEFF Research Database (Denmark)

Pedersen, Morten With

1997-01-01

The present thesis is about optimization of recurrent neural networks applied to time series modeling. In particular is considered fully recurrent networks working from only a single external input, one layer of nonlinear hidden units and a li near output unit applied to prediction of discrete time...... series. The overall objective s are to improve training by application of second-order methods and to improve generalization ability by architecture optimization accomplished by pruning. The major topics covered in the thesis are: 1. The problem of training recurrent networks is analyzed from a numerical...... of solution obtained as well as computation time required. 3. A theoretical definition of the generalization error for recurrent networks is provided. This definition justifies a commonly adopted approach for estimating generalization ability. 4. The viability of pruning recurrent networks by the Optimal...

Detection of Atrial Fibrillation Using Artifical Neural Network with Power Spectrum Density of RR Interval of Electrocardiogram

Science.gov (United States)

Afdala, Adfal; Nuryani, Nuryani; Satrio Nugroho, Anto

2017-01-01

Atrial fibrillation (AF) is a disorder of the heart with fairly high mortality in adults. AF is a common heart arrythmia which is characterized by a missing or irregular contraction of atria. Therefore, finding a method to detect atrial fibrillation is necessary. In this article a system to detect atrial fibrillation has been proposed. Detection system utilized backpropagation artifical neural network. Data input in this method includes power spectrum density of R-peaks interval of electrocardiogram which is selected by wrapping method. This research uses parameter learning rate, momentum, epoch and hidden layer. System produces good performance with accuracy, sensitivity, and specificity of 83.55%, 86.72 % and 81.47 %, respectively.

Robust Automatic Modulation Classification Technique for Fading Channels via Deep Neural Network

Directory of Open Access Journals (Sweden)

Jung Hwan Lee

2017-08-01

Full Text Available In this paper, we propose a deep neural network (DNN-based automatic modulation classification (AMC for digital communications. While conventional AMC techniques perform well for additive white Gaussian noise (AWGN channels, classification accuracy degrades for fading channels where the amplitude and phase of channel gain change in time. The key contributions of this paper are in two phases. First, we analyze the effectiveness of a variety of statistical features for AMC task in fading channels. We reveal that the features that are shown to be effective for fading channels are different from those known to be good for AWGN channels. Second, we introduce a new enhanced AMC technique based on DNN method. We use the extensive and diverse set of statistical features found in our study for the DNN-based classifier. The fully connected feedforward network with four hidden layers are trained to classify the modulation class for several fading scenarios. Numerical evaluation shows that the proposed technique offers significant performance gain over the existing AMC methods in fading channels.

Hidden Area and Mechanical Nonlinearities in Freestanding Graphene

Science.gov (United States)

Nicholl, Ryan J. T.; Lavrik, Nickolay V.; Vlassiouk, Ivan; Srijanto, Bernadeta R.; Bolotin, Kirill I.

2017-06-01

We investigated the effect of out-of-plane crumpling on the mechanical response of graphene membranes. In our experiments, stress was applied to graphene membranes using pressurized gas while the strain state was monitored through two complementary techniques: interferometric profilometry and Raman spectroscopy. By comparing the data obtained through these two techniques, we determined the geometric hidden area which quantifies the crumpling strength. While the devices with hidden area ˜0 % obeyed linear mechanics with biaxial stiffness 428 ±10 N /m , specimens with hidden area in the range 0.5%-1.0% were found to obey an anomalous nonlinear Hooke's law with an exponent ˜0.1 .

Pattern Recognition of Momentary Mental Workload Based on Multi-Channel Electrophysiological Data and Ensemble Convolutional Neural Networks.

Science.gov (United States)

Zhang, Jianhua; Li, Sunan; Wang, Rubin

2017-01-01

In this paper, we deal with the Mental Workload (MWL) classification problem based on the measured physiological data. First we discussed the optimal depth (i.e., the number of hidden layers) and parameter optimization algorithms for the Convolutional Neural Networks (CNN). The base CNNs designed were tested according to five classification performance indices, namely Accuracy, Precision, F-measure, G-mean, and required training time. Then we developed an Ensemble Convolutional Neural Network (ECNN) to enhance the accuracy and robustness of the individual CNN model. For the ECNN design, three model aggregation approaches (weighted averaging, majority voting and stacking) were examined and a resampling strategy was used to enhance the diversity of individual CNN models. The results of MWL classification performance comparison indicated that the proposed ECNN framework can effectively improve MWL classification performance and is featured by entirely automatic feature extraction and MWL classification, when compared with traditional machine learning methods.

Power level control of the TRIGA Mark-II research reactor using the multifeedback layer neural network and the particle swarm optimization

International Nuclear Information System (INIS)

Coban, Ramazan

2014-01-01

Highlights: • A multifeedback-layer neural network controller is presented for a research reactor. • Off-line learning of the MFLNN is accomplished by the PSO algorithm. • The results revealed that the MFLNN–PSO controller has a remarkable performance. - Abstract: In this paper, an artificial neural network controller is presented using the Multifeedback-Layer Neural Network (MFLNN), which is a recently proposed recurrent neural network, for neutronic power level control of a nuclear research reactor. Off-line learning of the MFLNN is accomplished by the Particle Swarm Optimization (PSO) algorithm. The MFLNN-PSO controller design is based on a nonlinear model of the TRIGA Mark-II research reactor. The learning and the test processes are implemented by means of a computer program at different power levels. The simulation results obtained reveal that the MFLNN-PSO controller has a remarkable performance on the neutronic power level control of the reactor for tracking the step reference power trajectories

Asymptotics for Estimating Equations in Hidden Markov Models

DEFF Research Database (Denmark)

Hansen, Jørgen Vinsløv; Jensen, Jens Ledet

Results on asymptotic normality for the maximum likelihood estimate in hidden Markov models are extended in two directions. The stationarity assumption is relaxed, which allows for a covariate process influencing the hidden Markov process. Furthermore a class of estimating equations is considered...

Two-layer anti-reflection strategies for implant applications

Science.gov (United States)

Guerrero, Douglas J.; Smith, Tamara; Kato, Masakazu; Kimura, Shigeo; Enomoto, Tomoyuki

2006-03-01

A two-layer bottom anti-reflective coating (BARC) concept in which a layer that develops slowly is coated on top of a bottom layer that develops more rapidly was demonstrated. Development rate control was achieved by selection of crosslinker amount and BARC curing conditions. A single-layer BARC was compared with the two-layer BARC concept. The single-layer BARC does not clear out of 200-nm deep vias. When the slower developing single-layer BARC was coated on top of the faster developing layer, the vias were cleared. Lithographic evaluation of the two-layer BARC concept shows the same resolution advantages as the single-layer system. Planarization properties of a two-layer BARC system are better than for a single-layer system, when comparing the same total nominal thicknesses.

A new method in prediction of TCP phases formation in superalloys

International Nuclear Information System (INIS)

Mousavi Anijdan, S.H.; Bahrami, A.

2005-01-01

The purpose of this investigation is to develop a model for prediction of topologically closed-packed (TCP) phases formation in superalloys. In this study, artificial neural networks (ANN), using several different network architectures, were used to investigate the complex relationships between TCP phases and chemical composition of superalloys. In order to develop an optimum ANN structure, more than 200 experimental data were used to train and test the neural network. The results of this investigation shows that a multilayer perceptron (MLP) form of the neural networks with one hidden layer and 10 nodes in the hidden layer has the lowest mean absolute error (MAE) and can be accurately used to predict the electron-hole number (N v ) and TCP phases formation in superalloys

Boltzmann learning of parameters in cellular neural networks

DEFF Research Database (Denmark)

Hansen, Lars Kai

1992-01-01

The use of Bayesian methods to design cellular neural networks for signal processing tasks and the Boltzmann machine learning rule for parameter estimation is discussed. The learning rule can be used for models with hidden units, or for completely unsupervised learning. The latter is exemplified...

Perspectives of intellectual processing of large volumes of astronomical data using neural networks

Science.gov (United States)

Gorbunov, A. A.; Isaev, E. A.; Samodurov, V. A.

2018-01-01

In the process of astronomical observations vast amounts of data are collected. BSA (Big Scanning Antenna) LPI used in the study of impulse phenomena, daily logs 87.5 GB of data (32 TB per year). This data has important implications for both short-and long-term monitoring of various classes of radio sources (including radio transients of different nature), monitoring the Earth’s ionosphere, the interplanetary and the interstellar plasma, the search and monitoring of different classes of radio sources. In the framework of the studies discovered 83096 individual pulse events (in the interval of the study highlighted July 2012 - October 2013), which may correspond to pulsars, twinkling springs, and a rapid radio transients. Detected impulse events are supposed to be used to filter subsequent observations. The study suggests approach, using the creation of the multilayered artificial neural network, which processes the input raw data and after processing, by the hidden layer, the output layer produces a class of impulsive phenomena.

Automatic detection of photoresist residual layer in lithography using a neural classification approach

KAUST Repository

Gereige, Issam

2012-09-01

Photolithography is a fundamental process in the semiconductor industry and it is considered as the key element towards extreme nanoscale integration. In this technique, a polymer photo sensitive mask with the desired patterns is created on the substrate to be etched. Roughly speaking, the areas to be etched are not covered with polymer. Thus, no residual layer should remain on these areas in order to insure an optimal transfer of the patterns on the substrate. In this paper, we propose a nondestructive method based on a classification approach achieved by artificial neural network for automatic residual layer detection from an ellipsometric signature. Only the case of regular defect, i.e. homogenous residual layer, will be considered. The limitation of the method will be discussed. Then, an experimental result on a 400 nm period grating manufactured with nanoimprint lithography is analyzed with our method. © 2012 Elsevier B.V. All rights reserved.

An interpretable LSTM neural network for autoregressive exogenous model

OpenAIRE

Guo, Tian; Lin, Tao; Lu, Yao

2018-01-01

In this paper, we propose an interpretable LSTM recurrent neural network, i.e., multi-variable LSTM for time series with exogenous variables. Currently, widely used attention mechanism in recurrent neural networks mostly focuses on the temporal aspect of data and falls short of characterizing variable importance. To this end, our multi-variable LSTM equipped with tensorized hidden states is developed to learn variable specific representations, which give rise to both temporal and variable lev...

Intelligent fault diagnosis of rolling bearings using an improved deep recurrent neural network

Science.gov (United States)

Jiang, Hongkai; Li, Xingqiu; Shao, Haidong; Zhao, Ke

2018-06-01

Traditional intelligent fault diagnosis methods for rolling bearings heavily depend on manual feature extraction and feature selection. For this purpose, an intelligent deep learning method, named the improved deep recurrent neural network (DRNN), is proposed in this paper. Firstly, frequency spectrum sequences are used as inputs to reduce the input size and ensure good robustness. Secondly, DRNN is constructed by the stacks of the recurrent hidden layer to automatically extract the features from the input spectrum sequences. Thirdly, an adaptive learning rate is adopted to improve the training performance of the constructed DRNN. The proposed method is verified with experimental rolling bearing data, and the results confirm that the proposed method is more effective than traditional intelligent fault diagnosis methods.

Estimating surface longwave radiative fluxes from satellites utilizing artificial neural networks

Science.gov (United States)

Nussbaumer, Eric A.; Pinker, Rachel T.

2012-04-01

A novel approach for calculating downwelling surface longwave (DSLW) radiation under all sky conditions is presented. The DSLW model (hereafter, DSLW/UMD v2) similarly to its predecessor, DSLW/UMD v1, is driven with a combination of Moderate Resolution Imaging Spectroradiometer (MODIS) level-3 cloud parameters and information from the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA-Interim model. To compute the clear sky component of DSLW a two layer feed-forward artificial neural network with sigmoid hidden neurons and linear output neurons is implemented; it is trained with simulations derived from runs of the Rapid Radiative Transfer Model (RRTM). When computing the cloud contribution to DSLW, the cloud base temperature is estimated by using an independent artificial neural network approach of similar architecture as previously mentioned, and parameterizations. The cloud base temperature neural network is trained using spatially and temporally co-located MODIS and CloudSat Cloud Profiling Radar (CPR) and the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) observations. Daily average estimates of DSLW from 2003 to 2009 are compared against ground measurements from the Baseline Surface Radiation Network (BSRN) giving an overall correlation coefficient of 0.98, root mean square error (rmse) of 15.84 W m-2, and a bias of -0.39 W m-2. This is an improvement over an earlier version of the model (DSLW/UMD v1) which for the same time period has an overall correlation coefficient 0.97 rmse of 17.27 W m-2, and bias of 0.73 W m-2.

Convective heat transfer and pressure drop of aqua based TiO2 nanofluids at different diameters of nanoparticles: Data analysis and modeling with artificial neural network

Science.gov (United States)

Hemmat Esfe, Mohammad; Nadooshan, Afshin Ahmadi; Arshi, Ali; Alirezaie, Ali

2018-03-01

In this study, experimental data related to the Nusselt number and pressure drop of aqueous nanofluids of Titania is modeled and estimated by using ANN with 2 hidden layers and 8 neurons in each layer. Also in this study the effect of various effective variables in the Nusselt number and pressure drop is surveyed. This study indicated that the neural network modeling has been able to model experimental data with great accuracy. The modeling regression coefficient for the data of Nusselt number and relative pressure drop is 99.94% and 99.97% respectively. Besides, it represented that the increment of the Reynolds number and concentration made the increment of Nusselt number and pressure drop of aqueous nanofluid.

Data Compression of Seismic Images by Neural Networks Compression d'images sismiques par des réseaux neuronaux

Directory of Open Access Journals (Sweden)

Epping W. J. M.

2006-11-01

Full Text Available Neural networks with the multi-layered perceptron architecture were trained on an autoassociation task to compress 2D seismic data. Networks with linear transfer functions outperformed nonlinear neural nets with single or multiple hidden layers. This indicates that the correlational structure of the seismic data is predominantly linear. A compression factor of 5 to 7 can be achieved if a reconstruction error of 10% is allowed. The performance on new test data was similar to that achieved with the training data. The hidden units developed feature-detecting properties that resemble oriented line, edge and more complex feature detectors. The feature detectors of linear neural nets are near-orthogonal rotations of the principal eigenvectors of the Karhunen-Loève transformation. Des réseaux neuronaux à architecture de perceptron multicouches ont été expérimentés en auto-association pour permettre la compression de données sismiques bidimensionnelles. Les réseaux neuronaux à fonctions de transfert linéaires s'avèrent plus performants que les réseaux neuronaux non linéaires, à une ou plusieurs couches cachées. Ceci indique que la structure corrélative des données sismiques est à prédominance linéaire. Un facteur de compression de 5 à 7 peut être obtenu si une erreur de reconstruction de 10 % est admise. La performance sur les données de test est très proche de celle obtenue sur les données d'apprentissage. Les unités cachées développent des propriétés de détection de caractéristiques ressemblant à des détecteurs de lignes orientées, de bords et de figures plus complexes. Les détecteurs de caractéristique des réseaux neuronaux linéaires sont des rotations quasi orthogonales des vecteurs propres principaux de la transformation de Karhunen-Loève.

Anomalous Signal Detection in ELF Band Electromagnetic Wave using Multi-layer Neural Network with Wavelet Decomposition

Science.gov (United States)

Itai, Akitoshi; Yasukawa, Hiroshi; Takumi, Ichi; Hata, Masayasu

It is well known that electromagnetic waves radiated from the earth's crust are useful for predicting earthquakes. We analyze the electromagnetic waves received at the extremely low frequency band of 223Hz. These observed signals contain the seismic radiation from the earth's crust, but also include several undesired signals. Our research focuses on the signal detection technique to identify an anomalous signal corresponding to the seismic radiation in the observed signal. Conventional anomalous signal detections lack a wide applicability due to their assumptions, e.g. the digital data have to be observed at the same time or the same sensor. In order to overcome the limitation related to the observed signal, we proposed the anomalous signals detection based on a multi-layer neural network which is trained by digital data observed during a span of a day. In the neural network approach, training data do not need to be recorded at the same place or the same time. However, some noises, which have a large amplitude, are detected as the anomalous signal. This paper develops a multi-layer neural network to decrease the false detection of the anomalous signal from the electromagnetic wave. The training data for the proposed network is the decomposed signal of the observed signal during several days, since the seismic radiations are often recorded from several days to a couple of weeks. Results show that the proposed neural network is useful to achieve the accurate detection of the anomalous signal that indicates seismic activity.

Neural network approximation of tip-abrasion effects in AFM imaging

International Nuclear Information System (INIS)

Bakucz, Peter; Dziomba, Thorsten; Koenders, Ludger; Krüger-Sehm, Rolf; Yacoot, Andrew

2008-01-01

The abrasion (wear) of tips used in scanning force microscopy (SFM) directly influences SFM image quality and is therefore of great relevance to quantitative SFM measurements. The increasing implementation of automated SFM measurement schemes has become a strong driving force for increasing efforts towards the prediction of tip wear, as it needs to be ensured that the probe is exchanged before a level of tip wear is reached that adversely affects the measurement quality. In this paper, we describe the identification of tip abrasion in a system of SFM measurements. We attempt to model the tip-abrasion process as a concatenation of a mapping from the measured AFM data to a regression vector and a nonlinear mapping from the regressor space to the output space. The mapping is formed as a basis function expansion. Feedforward neural networks are used to approximate this mapping. The one-hidden layer network gave a good quality of fit for the training and test sets for the tip-abrasion system. We illustrate our method with AFM measurements of both fine periodic structures and randomly oriented sharp features and compare our neural network results with those obtained using other methods

Neural network approximation of tip-abrasion effects in AFM imaging

Science.gov (United States)

Bakucz, Peter; Yacoot, Andrew; Dziomba, Thorsten; Koenders, Ludger; Krüger-Sehm, Rolf

2008-06-01

The abrasion (wear) of tips used in scanning force microscopy (SFM) directly influences SFM image quality and is therefore of great relevance to quantitative SFM measurements. The increasing implementation of automated SFM measurement schemes has become a strong driving force for increasing efforts towards the prediction of tip wear, as it needs to be ensured that the probe is exchanged before a level of tip wear is reached that adversely affects the measurement quality. In this paper, we describe the identification of tip abrasion in a system of SFM measurements. We attempt to model the tip-abrasion process as a concatenation of a mapping from the measured AFM data to a regression vector and a nonlinear mapping from the regressor space to the output space. The mapping is formed as a basis function expansion. Feedforward neural networks are used to approximate this mapping. The one-hidden layer network gave a good quality of fit for the training and test sets for the tip-abrasion system. We illustrate our method with AFM measurements of both fine periodic structures and randomly oriented sharp features and compare our neural network results with those obtained using other methods.

Prediction of degree of crystallinity for the LTA zeolite using artificial neural networks

Directory of Open Access Journals (Sweden)

Ghanbari Shahram

2017-10-01

Full Text Available Zeolites are microporous aluminosilicate/silicate crystalline materials with three-dimensional tetrahedral configuration. In this study, the degree of crystallinity of the synthesized Linde Type A (LTA zeolite, which is the main indicator of its quality/purity is tried to be modeled. Effect of crystallization time, temperature, molar ratio of the synthesis gel on the relative crystallinity of the LTA zeolites is investigated using artificial neural networks. Our experimental observations and some data collected from literature have been used for adjusting the parameters of the proposed model and evaluating its performance. It has been observed that two-layer perceptron network with eight hidden neurons is the most accurate approach for the considered task. The designed model predicts the experimental datasets with a mean square error of 3.99 × 10-6, absolute average relative deviation of 8.69 %, and regression coefficient of 0.9596. The proposed model can decrease the required time and number of experiments to evaluate the extent of crystallinity of the LTA zeolites.

A robust neural network-based approach for microseismic event detection

KAUST Repository

Akram, Jubran

2017-08-17

We present an artificial neural network based approach for robust event detection from low S/N waveforms. We use a feed-forward network with a single hidden layer that is tuned on a training dataset and later applied on the entire example dataset for event detection. The input features used include the average of absolute amplitudes, variance, energy-ratio and polarization rectilinearity. These features are calculated in a moving-window of same length for the entire waveform. The output is set as a user-specified relative probability curve, which provides a robust way of distinguishing between weak and strong events. An optimal network is selected by studying the weight-based saliency and effect of number of neurons on the predicted results. Using synthetic data examples, we demonstrate that this approach is effective in detecting weaker events and reduces the number of false positives.

Bosonization, dual transformation and non-local hidden symmetry in two dimensions

International Nuclear Information System (INIS)

Hata, Hiroyuki

1985-01-01

The non-local hidden symmetry is investigated in the bosonized non-abelian Thirring model and the dual representation of the chiral model. In these representations the first non-local symmetry is spontaneously broken in naive pertubation theory. (orig.)

Prediction of two-phase mixture density using artificial neural networks

International Nuclear Information System (INIS)

Lombardi, C.; Mazzola, A.

1997-01-01

In nuclear power plants, the density of boiling mixtures has a significant relevance due to its influence on the neutronic balance, the power distribution and the reactor dynamics. Since the determination of the two-phase mixture density on a purely analytical basis is in fact impractical in many situations of interest, heuristic relationships have been developed based on the parameters describing the two-phase system. However, the best or even a good structure for the correlation cannot be determined in advance, also considering that it is usually desired to represent the experimental data with the most compact equation. A possible alternative to empirical correlations is the use of artificial neural networks, which allow one to model complex systems without requiring the explicit formulation of the relationships existing among the variables. In this work, the neural network methodology was applied to predict the density data of two-phase mixtures up-flowing in adiabatic channels under different experimental conditions. The trained network predicts the density data with a root-mean-square error of 5.33%, being ∼ 93% of the data points predicted within 10%. When compared with those of two conventional well-proven correlations, i.e. the Zuber-Findlay and the CISE correlations, the neural network performances are significantly better. In spite of the good accuracy of the neural network predictions, the 'black-box' characteristic of the neural model does not allow an easy physical interpretation of the knowledge integrated in the network weights. Therefore, the neural network methodology has the advantage of not requiring a formal correlation structure and of giving very accurate results, but at the expense of a loss of model transparency. (author)

Local models and hidden nonlocality in Quantum Theory

OpenAIRE

Guerini, Leonardo

2014-01-01

This Master's thesis has two central subjects: the simulation of correlations generated by local measurements on entangled quantum states by local hidden-variables models and the revelation of hidden nonlocality. We present and detail the Werner's local model and the hidden nonlocality of some Werner states of dimension $d\\geq5$, the Gisin-Degorre's local model for a Werner state of dimension $d=2$ and the local model of Hirsch et al. for mixtures of the singlet state and noise, all of them f...

The MITLL NIST LRE 2015 Language Recognition System

Science.gov (United States)

2016-05-06

Cluster Target Classes Arabic Egyptian , Iraqi, Levantine, Maghrebi, Modern Standard Chinese Cantonese, Mandarin, Min, Wu English...42.69 Egyptian (ara-arz) 440 97.27 British English (eng-gbr) 47 0.51 Indian English (eng-sas) 418 7.82 American English (eng-usg) 428 100.37...are obtained by training a Deep Neural Network (DNN) using a seven hidden layer architecture . On these systems, all hidden layers have 1024 nodes

A Novel Approach to ECG Classification Based upon Two-Layered HMMs in Body Sensor Networks

Directory of Open Access Journals (Sweden)

Wei Liang

2014-03-01

Full Text Available This paper presents a novel approach to ECG signal filtering and classification. Unlike the traditional techniques which aim at collecting and processing the ECG signals with the patient being still, lying in bed in hospitals, our proposed algorithm is intentionally designed for monitoring and classifying the patient’s ECG signals in the free-living environment. The patients are equipped with wearable ambulatory devices the whole day, which facilitates the real-time heart attack detection. In ECG preprocessing, an integral-coefficient-band-stop (ICBS filter is applied, which omits time-consuming floating-point computations. In addition, two-layered Hidden Markov Models (HMMs are applied to achieve ECG feature extraction and classification. The periodic ECG waveforms are segmented into ISO intervals, P subwave, QRS complex and T subwave respectively in the first HMM layer where expert-annotation assisted Baum-Welch algorithm is utilized in HMM modeling. Then the corresponding interval features are selected and applied to categorize the ECG into normal type or abnormal type (PVC, APC in the second HMM layer. For verifying the effectiveness of our algorithm on abnormal signal detection, we have developed an ECG body sensor network (BSN platform, whereby real-time ECG signals are collected, transmitted, displayed and the corresponding classification outcomes are deduced and shown on the BSN screen.

A Fusion Face Recognition Approach Based on 7-Layer Deep Learning Neural Network

Directory of Open Access Journals (Sweden)

Jianzheng Liu

2016-01-01

Full Text Available This paper presents a method for recognizing human faces with facial expression. In the proposed approach, a motion history image (MHI is employed to get the features in an expressive face. The face can be seen as a kind of physiological characteristic of a human and the expressions are behavioral characteristics. We fused the 2D images of a face and MHIs which were generated from the same face’s image sequences with expression. Then the fusion features were used to feed a 7-layer deep learning neural network. The previous 6 layers of the whole network can be seen as an autoencoder network which can reduce the dimension of the fusion features. The last layer of the network can be seen as a softmax regression; we used it to get the identification decision. Experimental results demonstrated that our proposed method performs favorably against several state-of-the-art methods.

The neural network as a part of decision support system for quality management for production objects in machining process

Directory of Open Access Journals (Sweden)

Cherepanska I.Yu.

2017-04-01

Full Text Available The research discusses the use of artificial neural networks (ANN as components of a decision support system (DSS to automate quality control manufacturing facilities machining business at the production, which should be focused on the analysis of large amounts of heterogeneous information. The necessity to use ANN as a part of DSS is justified by the fact that quality control during production is multistage and time-consuming process that is formalized difficult, and moreover requires considerable information and material costs for the efficiency of manufacturing operations performed. Taking into account the existing experience of successful use of ANN to solve difficult formal problems associated with handling large volumes of diverse and rapidly changing information, the authors synthesized ANN for automated determination of the causes deterioration of the quality of production objects (PO in the performance of manufacturing operations application. Particular attention is paid to the definition of the dimension of the hidden layer ANN synthesized due to the fact that today still there is no analytical expression to determine the dimension of the hidden layer ANN and size of the latter is determined only by the experimental results of ANN several different structures by comparison the results, in particular the value of mean square error.

Identifikasi Penyakit Diabetes Millitus Menggunakan Jaringan Syaraf Tiruan Dengan Metode Perambatan-Balik (Backpropagation)

OpenAIRE

Sriyanto, -; Sutedi, -

2010-01-01

Diabetes Melitus (DM) is dangerous disease that affect many of the various layer of work society. This disease is not easy to accurately recognized by the general society. So we need to develop a system that can identify accurately. System is built using neural networks with backpropagation methods and the function activation sigmoid. Neural network architecture using 8 input layer, 2 output layer and 5 hidden layer. The results show that this methods succesfully clasifies data diabetics and ...

Identifikasi Penyakit Diabetes Millitus Menggunakan Jaringan Syaraf Tiruan Dengan Metode Perambatan-Balik (Backpropagation)

OpenAIRE

Sutedi, Sutedi

2018-01-01

Diabetes Melitus (DM) is dangerous disease that affect many of the various layer of work society. This disease is not easy to accurately recognized by the general society. So we need to develop a system that can identify accurately. System is built using neural networks with backpropagation methods and the function activation sigmoid. Neural network architecture using 8 input layer, 2 output layer and 5 hidden layer. The results show that this methods succesfully clasifies data diabetics and ...

Convolutional Neural Network for Histopathological Analysis of Osteosarcoma.

Science.gov (United States)

Mishra, Rashika; Daescu, Ovidiu; Leavey, Patrick; Rakheja, Dinesh; Sengupta, Anita

2018-03-01

Pathologists often deal with high complexity and sometimes disagreement over osteosarcoma tumor classification due to cellular heterogeneity in the dataset. Segmentation and classification of histology tissue in H&E stained tumor image datasets is a challenging task because of intra-class variations, inter-class similarity, crowded context, and noisy data. In recent years, deep learning approaches have led to encouraging results in breast cancer and prostate cancer analysis. In this article, we propose convolutional neural network (CNN) as a tool to improve efficiency and accuracy of osteosarcoma tumor classification into tumor classes (viable tumor, necrosis) versus nontumor. The proposed CNN architecture contains eight learned layers: three sets of stacked two convolutional layers interspersed with max pooling layers for feature extraction and two fully connected layers with data augmentation strategies to boost performance. The use of a neural network results in higher accuracy of average 92% for the classification. We compare the proposed architecture with three existing and proven CNN architectures for image classification: AlexNet, LeNet, and VGGNet. We also provide a pipeline to calculate percentage necrosis in a given whole slide image. We conclude that the use of neural networks can assure both high accuracy and efficiency in osteosarcoma classification.

Modularity and Sparsity: Evolution of Neural Net Controllers in Physically Embodied Robots

Directory of Open Access Journals (Sweden)

Nicholas Livingston

2016-12-01

Full Text Available While modularity is thought to be central for the evolution of complexity and evolvability, it remains unclear how systems boot-strap themselves into modularity from random or fully integrated starting conditions. Clune et al. (2013 suggested that a positive correlation between sparsity and modularity is the prime cause of this transition. We sought to test the generality of this modularity-sparsity hypothesis by testing it for the first time in physically embodied robots. A population of ten Tadros — autonomous, surface-swimming robots propelled by a flapping tail — was used. Individuals varied only in the structure of their neural net control, a 2 x 6 x 2 network with recurrence in the hidden layer. Each of the 60 possible connections was coded in the genome, and could achieve one of three states: -1, 0, 1. Inputs were two light-dependent resistors and outputs were two motor control variables to the flapping tail, one for the frequency of the flapping and the other for the turning offset. Each Tadro was tested separately in a circular tank lit by a single overhead light source. Fitness was the amount of light gathered by a vertically oriented sensor that was disconnected from the controller net. Reproduction was asexual, with the top performer cloned and then all individuals entered into a roulette wheel selection process, with genomes mutated to create the offspring. The starting population of networks was randomly generated. Over ten generations, the population’s mean fitness increased two-fold. This evolution occurred in spite of an unintentional integer overflow problem in recurrent nodes in the hidden layer that caused outputs to oscillate. Our investigation of the oscillatory behavior showed that the mutual information of inputs and outputs was sufficient for the reactive behaviors observed. While we had predicted that both modularity and sparsity would follow the same trend as fitness, neither did so. Instead, selection gradients

Comparing success levels of different neural network structures in extracting discriminative information from the response patterns of a temperature-modulated resistive gas sensor

Science.gov (United States)

Hosseini-Golgoo, S. M.; Bozorgi, H.; Saberkari, A.

2015-06-01

Performances of three neural networks, consisting of a multi-layer perceptron, a radial basis function, and a neuro-fuzzy network with local linear model tree training algorithm, in modeling and extracting discriminative features from the response patterns of a temperature-modulated resistive gas sensor are quantitatively compared. For response pattern recording, a voltage staircase containing five steps each with a 20 s plateau is applied to the micro-heater of the sensor, when 12 different target gases, each at 11 concentration levels, are present. In each test, the hidden layer neuron weights are taken as the discriminatory feature vector of the target gas. These vectors are then mapped to a 3D feature space using linear discriminant analysis. The discriminative information content of the feature vectors are determined by the calculation of the Fisher’s discriminant ratio, affording quantitative comparison among the success rates achieved by the different neural network structures. The results demonstrate a superior discrimination ratio for features extracted from local linear neuro-fuzzy and radial-basis-function networks with recognition rates of 96.27% and 90.74%, respectively.

Comparing success levels of different neural network structures in extracting discriminative information from the response patterns of a temperature-modulated resistive gas sensor

International Nuclear Information System (INIS)

Hosseini-Golgoo, S M; Bozorgi, H; Saberkari, A

2015-01-01

Performances of three neural networks, consisting of a multi-layer perceptron, a radial basis function, and a neuro-fuzzy network with local linear model tree training algorithm, in modeling and extracting discriminative features from the response patterns of a temperature-modulated resistive gas sensor are quantitatively compared. For response pattern recording, a voltage staircase containing five steps each with a 20 s plateau is applied to the micro-heater of the sensor, when 12 different target gases, each at 11 concentration levels, are present. In each test, the hidden layer neuron weights are taken as the discriminatory feature vector of the target gas. These vectors are then mapped to a 3D feature space using linear discriminant analysis. The discriminative information content of the feature vectors are determined by the calculation of the Fisher’s discriminant ratio, affording quantitative comparison among the success rates achieved by the different neural network structures. The results demonstrate a superior discrimination ratio for features extracted from local linear neuro-fuzzy and radial-basis-function networks with recognition rates of 96.27% and 90.74%, respectively. (paper)

Two-Stage Hidden Markov Model in Gesture Recognition for Human Robot Interaction

Directory of Open Access Journals (Sweden)

Nhan Nguyen-Duc-Thanh

2012-07-01

Full Text Available Hidden Markov Model (HMM is very rich in mathematical structure and hence can form the theoretical basis for use in a wide range of applications including gesture representation. Most research in this field, however, uses only HMM for recognizing simple gestures, while HMM can definitely be applied for whole gesture meaning recognition. This is very effectively applicable in Human-Robot Interaction (HRI. In this paper, we introduce an approach for HRI in which not only the human can naturally control the robot by hand gesture, but also the robot can recognize what kind of task it is executing. The main idea behind this method is the 2-stages Hidden Markov Model. The 1st HMM is to recognize the prime command-like gestures. Based on the sequence of prime gestures that are recognized from the 1st stage and which represent the whole action, the 2nd HMM plays a role in task recognition. Another contribution of this paper is that we use the output Mixed Gaussian distribution in HMM to improve the recognition rate. In the experiment, we also complete a comparison of the different number of hidden states and mixture components to obtain the optimal one, and compare to other methods to evaluate this performance.

neural network based model o work based model of an industrial oil

African Journals Online (AJOL)

eobe

technique. g, Neural Network Model, Regression, Mean Square Error, PID controller. ... during the training processes. An additio ... used to carry out simulation studies of the mode .... A two-layer feed-forward neural network with Matlab.

Neural network modelling of planform geometry of headland-bay beaches

Science.gov (United States)

Iglesias, G.; López, I.; Castro, A.; Carballo, R.

2009-02-01

The shoreline of beaches in the lee of coastal salients or man-made structures, usually known as headland-bay beaches, has a distinctive curvature; wave fronts curve as a result of wave diffraction at the headland and in turn cause the shoreline to bend. The ensuing curved planform is of great interest both as a peculiar landform and in the context of engineering projects in which it is necessary to predict how a coastal structure will affect the sandy shoreline in its lee. A number of empirical models have been put forward, each based on a specific equation. A novel approach, based on the application of artificial neural networks, is presented in this work. Unlike the conventional method, no particular equation of the planform is embedded in the model. Instead, it is the model itself that learns about the problem from a series of examples of headland-bay beaches (the training set) and thereafter applies this self-acquired knowledge to other cases (the test set) for validation. Twenty-three headland-bay beaches from around the world were selected, of which sixteen and seven make up the training and test sets, respectively. As there is no well-developed theory for deciding upon the most convenient neural network architecture to deal with a particular data set, an experimental study was conducted in which ten different architectures with one and two hidden neuron layers and five training algorithms - 50 different options combining network architecture and training algorithm - were compared. Each of these options was implemented, trained and tested in order to find the best-performing approach for modelling the planform of headland-bay beaches. Finally, the selected neural network model was compared with a state-of-the-art planform model and was shown to outperform it.

Inductive differentiation of two neural lineages reconstituted in a microculture system from Xenopus early gastrula cells.

Science.gov (United States)

Mitani, S; Okamoto, H

1991-05-01

Neural induction of ectoderm cells has been reconstituted and examined in a microculture system derived from dissociated early gastrula cells of Xenopus laevis. We have used monoclonal antibodies as specific markers to monitor cellular differentiation from three distinct ectoderm lineages in culture (N1 for CNS neurons from neural tube, Me1 for melanophores from neural crest and E3 for skin epidermal cells from epidermal lineages). CNS neurons and melanophores differentiate when deep layer cells of the ventral ectoderm (VE, prospective epidermis region; 150 cells/culture) and an appropriate region of the marginal zone (MZ, prospective mesoderm region; 5-150 cells/culture) are co-cultured, but not in cultures of either cell type on their own; VE cells cultured alone yield epidermal cells as we have previously reported. The extent of inductive neural differentiation in the co-culture system strongly depends on the origin and number of MZ cells initially added to culture wells. The potency to induce CNS neurons is highest for dorsal MZ cells and sharply decreases as more ventrally located cells are used. The same dorsoventral distribution of potency is seen in the ability of MZ cells to inhibit epidermal differentiation. In contrast, the ability of MZ cells to induce melanophores shows the reverse polarity, ventral to dorsal. These data indicate that separate developmental mechanisms are used for the induction of neural tube and neural crest lineages. Co-differentiation of CNS neurons or melanophores with epidermal cells can be obtained in a single well of co-cultures of VE cells (150) and a wide range of numbers of MZ cells (5 to 100). Further, reproducible differentiation of both neural lineages requires intimate association between cells from the two gastrula regions; virtually no differentiation is obtained when cells from the VE and MZ are separated in a culture well. These results indicate that the inducing signals from MZ cells for both neural tube and neural

Applying deep neural networks to HEP job classification

International Nuclear Information System (INIS)

Wang, L; Shi, J; Yan, X

2015-01-01

The cluster of IHEP computing center is a middle-sized computing system which provides 10 thousands CPU cores, 5 PB disk storage, and 40 GB/s IO throughput. Its 1000+ users come from a variety of HEP experiments. In such a system, job classification is an indispensable task. Although experienced administrator can classify a HEP job by its IO pattern, it is unpractical to classify millions of jobs manually. We present how to solve this problem with deep neural networks in a supervised learning way. Firstly, we built a training data set of 320K samples by an IO pattern collection agent and a semi-automatic process of sample labelling. Then we implemented and trained DNNs models with Torch. During the process of model training, several meta-parameters was tuned with cross-validations. Test results show that a 5- hidden-layer DNNs model achieves 96% precision on the classification task. By comparison, it outperforms a linear model by 8% precision. (paper)

Artificial neural networks and adaptive neuro-fuzzy assessments for ground-coupled heat pump system

Energy Technology Data Exchange (ETDEWEB)

Esen, Hikmet; Esen, Mehmet [Department of Mechanical Education, Faculty of Technical Education, Firat University, 23119 Elazig (Turkey); Inalli, Mustafa [Department of Mechanical Engineering, Faculty of Engineering, Firat University, 23279 Elazig (Turkey); Sengur, Abdulkadir [Department of Electronic and Computer Science, Faculty of Technical Education, Firat University, 23119 Elazig (Turkey)

2008-07-01

This article present a comparison of artificial neural network (ANN) and adaptive neuro-fuzzy inference systems (ANFIS) applied for modelling a ground-coupled heat pump system (GCHP). The aim of this study is predicting system performance related to ground and air (condenser inlet and outlet) temperatures by using desired models. Performance forecasting is the precondition for the optimal design and energy-saving operation of air-conditioning systems. So obtained models will help the system designer to realize this precondition. The most suitable algorithm and neuron number in the hidden layer are found as Levenberg-Marquardt (LM) with seven neurons for ANN model whereas the most suitable membership function and number of membership functions are found as Gauss and two, respectively, for ANFIS model. The root-mean squared (RMS) value and the coefficient of variation in percent (cov) value are 0.0047 and 0.1363, respectively. The absolute fraction of variance (R{sup 2}) is 0.9999 which can be considered as very promising. This paper shows the appropriateness of ANFIS for the quantitative modeling of GCHP systems. (author)

Automatic target recognition using a feature-based optical neural network

Science.gov (United States)

Chao, Tien-Hsin

1992-01-01

An optical neural network based upon the Neocognitron paradigm (K. Fukushima et al. 1983) is introduced. A novel aspect of the architectural design is shift-invariant multichannel Fourier optical correlation within each processing layer. Multilayer processing is achieved by iteratively feeding back the output of the feature correlator to the input spatial light modulator and updating the Fourier filters. By training the neural net with characteristic features extracted from the target images, successful pattern recognition with intra-class fault tolerance and inter-class discrimination is achieved. A detailed system description is provided. Experimental demonstration of a two-layer neural network for space objects discrimination is also presented.

ESTUDIO DE SERIES TEMPORALES DE CONTAMINACIÓN AMBIENTAL MEDIANTE TÉCNICAS DE REDES NEURONALES ARTIFICIALES TIME SERIES ANALYSIS OF ATMOSPHERE POLLUTION DATA USING ARTIFICIAL NEURAL NETWORKS TECHNIQUES

Directory of Open Access Journals (Sweden)

Giovanni Salini Calderón

2006-12-01

concentrations between May and August for years between 1994 and 1996. In order to find the optimal time spacing between data and the number of values into the past necessary to forecast a future value, two standard tests were performed, Average Mutual Information (AMI and False Nearest Neighbours (FNN. The results of these tests suggest that the most convenient choice for modelling was to use 4 data with 6 hour spacing on a given day as input in order to forecast the value at 6 AM on the following day. Once the number and type of input and output variables are fixed, we implemented a forecasting model based on the neural network technique. We used a feedforward multilayer neural network and we trained it with the backpropagation algorithm. We tested networks with none, one and two hidden layers. The best model was one with one hidden layer, in contradiction with a previous study that found that minimum error was obtained with a net without hidden layer. Forecasts with the neural network are more accurate than those produced with a persistence model (the value six hours ahead is the same as the actual value.

Cough event classification by pretrained deep neural network.

Science.gov (United States)

Liu, Jia-Ming; You, Mingyu; Wang, Zheng; Li, Guo-Zheng; Xu, Xianghuai; Qiu, Zhongmin

2015-01-01

Cough is an essential symptom in respiratory diseases. In the measurement of cough severity, an accurate and objective cough monitor is expected by respiratory disease society. This paper aims to introduce a better performed algorithm, pretrained deep neural network (DNN), to the cough classification problem, which is a key step in the cough monitor. The deep neural network models are built from two steps, pretrain and fine-tuning, followed by a Hidden Markov Model (HMM) decoder to capture tamporal information of the audio signals. By unsupervised pretraining a deep belief network, a good initialization for a deep neural network is learned. Then the fine-tuning step is a back propogation tuning the neural network so that it can predict the observation probability associated with each HMM states, where the HMM states are originally achieved by force-alignment with a Gaussian Mixture Model Hidden Markov Model (GMM-HMM) on the training samples. Three cough HMMs and one noncough HMM are employed to model coughs and noncoughs respectively. The final decision is made based on viterbi decoding algorihtm that generates the most likely HMM sequence for each sample. A sample is labeled as cough if a cough HMM is found in the sequence. The experiments were conducted on a dataset that was collected from 22 patients with respiratory diseases. Patient dependent (PD) and patient independent (PI) experimental settings were used to evaluate the models. Five criteria, sensitivity, specificity, F1, macro average and micro average are shown to depict different aspects of the models. From overall evaluation criteria, the DNN based methods are superior to traditional GMM-HMM based method on F1 and micro average with maximal 14% and 11% error reduction in PD and 7% and 10% in PI, meanwhile keep similar performances on macro average. They also surpass GMM-HMM model on specificity with maximal 14% error reduction on both PD and PI. In this paper, we tried pretrained deep neural network in

Hidden U (1 ) gauge symmetry realizing a neutrinophilic two-Higgs-doublet model with dark matter

Science.gov (United States)

Nomura, Takaaki; Okada, Hiroshi

2018-04-01

We propose a neutrinophilic two-Higgs-doublet model with hidden local U (1 ) symmetry, where active neutrinos are Dirac type, and a fermionic dark matter (DM) candidate is naturally induced as a result of remnant symmetry even after the spontaneous symmetry breaking. In addition, a physical Goldstone boson arises as a consequence of two types of gauge singlet bosons and contributes to the DM phenomenologies as well as an additional neutral gauge boson. Then, we analyze the relic density of DM within the safe range of direct detection searches and show the allowed region of dark matter mass.

Combining neural networks and genetic algorithms for hydrological flow forecasting

Science.gov (United States)

Neruda, Roman; Srejber, Jan; Neruda, Martin; Pascenko, Petr

2010-05-01

We present a neural network approach to rainfall-runoff modeling for small size river basins based on several time series of hourly measured data. Different neural networks are considered for short time runoff predictions (from one to six hours lead time) based on runoff and rainfall data observed in previous time steps. Correlation analysis shows that runoff data, short time rainfall history, and aggregated API values are the most significant data for the prediction. Neural models of multilayer perceptron and radial basis function networks with different numbers of units are used and compared with more traditional linear time series predictors. Out of possible 48 hours of relevant history of all the input variables, the most important ones are selected by means of input filters created by a genetic algorithm. The genetic algorithm works with population of binary encoded vectors defining input selection patterns. Standard genetic operators of two-point crossover, random bit-flipping mutation, and tournament selection were used. The evaluation of objective function of each individual consists of several rounds of building and testing a particular neural network model. The whole procedure is rather computational exacting (taking hours to days on a desktop PC), thus a high-performance mainframe computer has been used for our experiments. Results based on two years worth data from the Ploucnice river in Northern Bohemia suggest that main problems connected with this approach to modeling are ovetraining that can lead to poor generalization, and relatively small number of extreme events which makes it difficult for a model to predict the amplitude of the event. Thus, experiments with both absolute and relative runoff predictions were carried out. In general it can be concluded that the neural models show about 5 per cent improvement in terms of efficiency coefficient over liner models. Multilayer perceptrons with one hidden layer trained by back propagation algorithm and

Hidden photons in beam dump experiments and in connection with dark matter

Energy Technology Data Exchange (ETDEWEB)

Andreas, Sarah

2012-12-15

Hidden sectors with light extra U(1) gauge bosons, so-called hidden photons, recently received much interest as natural feature of beyond standard model scenarios like string theory and SUSY and because of their possible connection to dark matter. This paper presents limits on hidden photons from past electron beam dump experiments including two new limits from experiments at KEK and Orsay. Additionally, various hidden sector models containing both a hidden photon and a dark matter candidate are discussed with respect to their viability and potential signatures in direct detection.

Hidden photons in beam dump experiments and in connection with dark matter

International Nuclear Information System (INIS)

Andreas, Sarah

2012-12-01

Hidden sectors with light extra U(1) gauge bosons, so-called hidden photons, recently received much interest as natural feature of beyond standard model scenarios like string theory and SUSY and because of their possible connection to dark matter. This paper presents limits on hidden photons from past electron beam dump experiments including two new limits from experiments at KEK and Orsay. Additionally, various hidden sector models containing both a hidden photon and a dark matter candidate are discussed with respect to their viability and potential signatures in direct detection.

Estimation of scattered photons using a neural network in SPECT

International Nuclear Information System (INIS)

Hasegawa, Wataru; Ogawa, Koichi

1994-01-01

In single photon emission CT (SPECT), measured projection data involve scattered photons. This causes degradation of spatial resolution and contrast in reconstructed images. The purpose of this study is to estimate the scattered photons, and eliminate them from measured data. To estimate the scattered photons, we used an artificial neural network which consists of five input units, five hidden units, and two output units. The inputs of the network are the ratios of the counts acquired by five narrow energy windows and their sum. The outputs are the ratios of the count of scattered photons and that of primary photons to the total count. The neural network was trained with a back-propagation algorithm using count data obtained by a Monte Carlo simulation. The results of simulation showed improvement of contrast and spatial resolution in reconstructed images. (author)

Distinguishing Hidden Markov Chains

OpenAIRE

Kiefer, Stefan; Sistla, A. Prasad

2015-01-01

Hidden Markov Chains (HMCs) are commonly used mathematical models of probabilistic systems. They are employed in various fields such as speech recognition, signal processing, and biological sequence analysis. We consider the problem of distinguishing two given HMCs based on an observation sequence that one of the HMCs generates. More precisely, given two HMCs and an observation sequence, a distinguishing algorithm is expected to identify the HMC that generates the observation sequence. Two HM...

Strength of the Three Layer Beam with Two Binding Layers

Directory of Open Access Journals (Sweden)

Smyczyński M. J.

2016-09-01

Full Text Available The paper is devoted to the strength analysis of a simply supported three layer beam. The sandwich beam consists of: two metal facings, the metal foam core and two binding layers between the faces and the core. In consequence, the beam is a five layer beam. The main goal of the study is to elaborate a mathematical model of this beam, analytical description and a solution of the three-point bending problem. The beam is subjected to a transverse load. The nonlinear hypothesis of the deformation of the cross section of the beam is formulated. Based on the principle of the stationary potential energy the system of four equations of equilibrium is derived. Then deflections and stresses are determined. The influence of the binding layers is considered. The results of the solutions of the bending problem analysis are shown in the tables and figures. The analytical model is verified numerically using the finite element analysis, as well as experimentally.

SpineCreator: a Graphical User Interface for the Creation of Layered Neural Models.

Science.gov (United States)

Cope, A J; Richmond, P; James, S S; Gurney, K; Allerton, D J

2017-01-01

There is a growing requirement in computational neuroscience for tools that permit collaborative model building, model sharing, combining existing models into a larger system (multi-scale model integration), and are able to simulate models using a variety of simulation engines and hardware platforms. Layered XML model specification formats solve many of these problems, however they are difficult to write and visualise without tools. Here we describe a new graphical software tool, SpineCreator, which facilitates the creation and visualisation of layered models of point spiking neurons or rate coded neurons without requiring the need for programming. We demonstrate the tool through the reproduction and visualisation of published models and show simulation results using code generation interfaced directly into SpineCreator. As a unique application for the graphical creation of neural networks, SpineCreator represents an important step forward for neuronal modelling.

Predicting methionine and lysine contents in soybean meal and fish meal using a group method of data handling-type neural network

Energy Technology Data Exchange (ETDEWEB)

Mottaghitalab, M.; Nikkhah, N.; Darmani-Kuhi, H.; López, S.; France, J.

2015-07-01

Artificial neural network models offer an alternative to linear regression analysis for predicting the amino acid content of feeds from their chemical composition. A group method of data handling-type neural network (GMDH-type NN), with an evolutionary method of genetic algorithm, was used to predict methionine (Met) and lysine (Lys) contents of soybean meal (SBM) and fish meal (FM) from their proximate analyses (i.e. crude protein, crude fat, crude fibre, ash and moisture). A data set with 119 data lines for Met and 116 lines for Lys was used to develop GMDH-type NN models with two hidden layers. The data lines were divided into two groups to produce training and validation sets. The data sets were imported into the GEvoM software for training the networks. The predictive capability of the constructed models was evaluated by their abilities to estimate the validation data sets accurately. A quantitative examination of goodness of fit for the predictive models was made using a number of precision, concordance and bias statistics. The statistical performance of the models developed revealed close agreement between observed and predicted Met and Lys contents for SBM and FM. The results of this study clearly illustrate the validity of GMDH-type NN models to estimate accurately the amino acid content of poultry feed ingredients from their chemical composition . (Author)

Forced phase-locked states and information retrieval in a two-layer network of oscillatory neurons with directional connectivity

International Nuclear Information System (INIS)

Kazantsev, Victor; Pimashkin, Alexey

2007-01-01

We propose two-layer architecture of associative memory oscillatory network with directional interlayer connectivity. The network is capable to store information in the form of phase-locked (in-phase and antiphase) oscillatory patterns. The first (input) layer takes an input pattern to be recognized and their units are unidirectionally connected with all units of the second (control) layer. The connection strengths are weighted using the Hebbian rule. The output (retrieved) patterns appear as forced-phase locked states of the control layer. The conditions are found and analytically expressed for pattern retrieval in response on incoming stimulus. It is shown that the system is capable to recover patterns with a certain level of distortions or noises in their profiles. The architecture is implemented with the Kuramoto phase model and using synaptically coupled neural oscillators with spikes. It is found that the spiking model is capable to retrieve patterns using the spiking phase that translates memorized patterns into the spiking phase shifts at different time scales

MODELADO DEL PRECIO SPOT DE LA ELECTRICIDAD EN BRASIL USANDO UNA RED NEURONAL AUTORREGRESIVA ELECTRICITY SPOT PRICE MODELLING IN BRASIL USING AN AUTOREGRESSIVE NEURAL NETWORK

Directory of Open Access Journals (Sweden)

Juan D Velásquez

2008-12-01

Full Text Available Una red neuronal autorregresiva es estimada para el precio mensual brasileño de corto plazo de la electricidad, la cual describe mejor la dinámica de los precios que un modelo lineal autorregresivo y que un perceptrón multicapa clásico que usan las mismas entradas y neuronas en la capa oculta. El modelo propuesto es especificado usando un procedimiento estadístico basado en el contraste del radio de verosimilitud. El modelo pasa una batería de pruebas de diagnóstico. El procedimiento de especificación propuesto permite seleccionar el número de unidades en la capa oculta y las entradas a la red neuronal, usando pruebas estadísticas que tienen en cuenta la cantidad de los datos y el ajuste del modelo a la serie de precios. La especificación del modelo final demuestra que el precio para el próximo mes es una función no lineal del precio actual, de la energía afluente actual y de la energía almacenada en el embalse equivalente en el mes actual y dos meses atrás.An autoregressive neural network model is estimated for the monthly Brazilian electricity spot price, which describes the prices dynamics better than a linear autoregressive model and a classical multilayer perceptron using the same input and neurons in the hidden layer. The proposed model is specified using a statistical procedure based on a likelihood ratio test. The model passes a battery of diagnostic tests. The proposed specification procedure allows us to select the number of units in hidden layer and the inputs to the neural network based on statistical tests, taking into account the number of data and the model fitting to the price time series. The final model specification demonstrates that the price for the next month is a nonlinear function of the current price, the current energy inflow, and the energy saved in the equivalent reservoir in the current month and two months ago.

A novel framework for intelligent signal detection via artificial neural networks for cyclic voltammetry in pyroprocessing technology

International Nuclear Information System (INIS)

Rakhshan Pouri, Samaneh; Manic, Milos; Phongikaroon, Supathorn

2018-01-01

Highlights: •First time ANN implementation toward pyroprocessing safeguards. •Real time monitoring in terms of intelligent materials detection and accountability. •CV simulation via ANN showing a high accuracy of prediction for the unseen situation. •Elimination of trial and error approach to avoid overfitting in learning. -- Abstract: Electrorefiner (ER) is the heart of pyroprocessing technology which contains different fission, rare-earth, and transuranic chloride compositions during the operation. This is still a developing technology that needs to be advanced for the commercial reprocessing design of used nuclear fuel (UNF) in terms of intelligent materials detection and accountability towards safeguards. A novel signal detection, artificial neural network (ANN), has been proposed in this study to apply on massive ER systemic parameters to simulate cyclic voltammetry (CV) graphs for the unseen situation. ANN could be trained to mimic the system by driving the data sets interrelation between variables to provide current and potential simulated data sets with a high accuracy of prediction. For this purpose, over 230,000 experimental data points reported in literature have been explored—0.5–5 wt% of zirconium chloride (ZrCl 4 ) in LiCl-KCl molten salt with different scan rates at 773 K. This study has illustrated a new framework of ANN implementation to eliminate trial and error approach by comparing the average error of one to three hidden layers with different number of neurons. In addition, this framework results in finding a preferable balance between underfitting and overfitting in deep learning. Furthermore, simulated CV graphs were compared with the experimental data and illustrated a reasonable prediction. The results reveal two structures with three hidden layers providing a good prediction with a low average error. The outcomes indicate that ANN has a strong potential in applying toward safeguards for pyroprocessing technology.

NETS - A NEURAL NETWORK DEVELOPMENT TOOL, VERSION 3.0 (MACINTOSH VERSION)

Science.gov (United States)

Phillips, T. A.

1994-01-01

NETS, A Tool for the Development and Evaluation of Neural Networks, provides a simulation of Neural Network algorithms plus an environment for developing such algorithms. Neural Networks are a class of systems modeled after the human brain. Artificial Neural Networks are formed from hundreds or thousands of simulated neurons, connected to each other in a manner similar to brain neurons. Problems which involve pattern matching readily fit the class of problems which NETS is designed to solve. NETS uses the back propagation learning method for all of the networks which it creates. The nodes of a network are usually grouped together into clumps called layers. Generally, a network will have an input layer through which the various environment stimuli are presented to the network, and an output layer for determining the network's response. The number of nodes in these two layers is usually tied to some features of the problem being solved. Other layers, which form intermediate stops between the input and output layers, are called hidden layers. NETS allows the user to customize the patterns of connections between layers of a network. NETS also provides features for saving the weight values of a network during the learning process, which allows for more precise control over the learning process. NETS is an interpreter. Its method of execution is the familiar "read-evaluate-print" loop found in interpreted languages such as BASIC and LISP. The user is presented with a prompt which is the simulator's way of asking for input. After a command is issued, NETS will attempt to evaluate the command, which may produce more prompts requesting specific information or an error if the command is not understood. The typical process involved when using NETS consists of translating the problem into a format which uses input/output pairs, designing a network configuration for the problem, and finally training the network with input/output pairs until an acceptable error is reached. NETS

Hidden gauge symmetry

International Nuclear Information System (INIS)

O'Raifeartaigh, L.

1979-01-01

This review describes the principles of hidden gauge symmetry and of its application to the fundamental interactions. The emphasis is on the structure of the theory rather than on the technical details and, in order to emphasise the structure, gauge symmetry and hidden symmetry are first treated as independent phenomena before being combined into a single (hidden gauge symmetric) theory. The main application of the theory is to the weak and electromagnetic interactions of the elementary particles, and although models are used for comparison with experiment and for illustration, emphasis is placed on those features of the application which are model-independent. (author)

An electronic system for simulation of neural networks with a micro-second real time constraint

International Nuclear Information System (INIS)

Chorti, Arsenia; Granado, Bertrand; Denby, Bruce; Garda, Patrick

2001-01-01

Neural networks implemented in hardware can perform pattern recognition very quickly, and as such have been used to advantage in the triggering systems of certain high energy physics experiments. Typically, time constants of the order of a few microseconds are required. In this paper, we present a new system. MAHARADJA, for evaluating MLP and RBF neural network paradigms in real time. The system is tested on a possible ATLAS muon triggering application suggested by the Tel Aviv ATLAS group, consisting of a 4-8-8-4 MLP which must be evaluated in 10 microseconds. The inputs to the net are dx/dz, x(z=0), dy/dz, and y(z=0), whereas the outputs give pt, tan(phi), sin(theta), and q, the charge. With a 10 MHz clock, MAHARADJA calculates the result in 6.8 microseconds; at 20 MHz, which is readily attainable, this would be reduced to only 3.4 microseconds. The system can also handle RBF networks with 3 different distance metrics (Euclidean, Manhattan and Mahalanobis), and can simulate any MLP of 10 hidden layers or less. The electronic implementation is with FPGA's, which can be optimized for a specific neural network because the number of processing elements can be modified

Clinical Outcome Prediction in Aneurysmal Subarachnoid Hemorrhage Using Bayesian Neural Networks with Fuzzy Logic Inferences

Directory of Open Access Journals (Sweden)

Benjamin W. Y. Lo

2013-01-01

Full Text Available Objective. The novel clinical prediction approach of Bayesian neural networks with fuzzy logic inferences is created and applied to derive prognostic decision rules in cerebral aneurysmal subarachnoid hemorrhage (aSAH. Methods. The approach of Bayesian neural networks with fuzzy logic inferences was applied to data from five trials of Tirilazad for aneurysmal subarachnoid hemorrhage (3551 patients. Results. Bayesian meta-analyses of observational studies on aSAH prognostic factors gave generalizable posterior distributions of population mean log odd ratios (ORs. Similar trends were noted in Bayesian and linear regression ORs. Significant outcome predictors include normal motor response, cerebral infarction, history of myocardial infarction, cerebral edema, history of diabetes mellitus, fever on day 8, prior subarachnoid hemorrhage, admission angiographic vasospasm, neurological grade, intraventricular hemorrhage, ruptured aneurysm size, history of hypertension, vasospasm day, age and mean arterial pressure. Heteroscedasticity was present in the nontransformed dataset. Artificial neural networks found nonlinear relationships with 11 hidden variables in 1 layer, using the multilayer perceptron model. Fuzzy logic decision rules (centroid defuzzification technique denoted cut-off points for poor prognosis at greater than 2.5 clusters. Discussion. This aSAH prognostic system makes use of existing knowledge, recognizes unknown areas, incorporates one's clinical reasoning, and compensates for uncertainty in prognostication.

Artificial Neural Network Modelling of the Energy Content of Municipal Solid Wastes in Northern Nigeria

Directory of Open Access Journals (Sweden)

M. B. Oumarou

2017-12-01

Full Text Available The study presents an application of the artificial neural network model using the back propagation learning algorithm to predict the actual calorific value of the municipal solid waste in major cities of the northern part of Nigeria, with high population densities and intense industrial activities. These cities are: Kano, Damaturu, Dutse, Bauchi, Birnin Kebbi, Gusau, Maiduguri, Katsina and Sokoto. Experimental data of the energy content and the physical characterization of the municipal solid waste serve as the input parameter in nature of wood, grass, metal, plastic, food remnants, leaves, glass and paper. Comparative studies were made by using the developed model, the experimental results and a correlation which was earlier developed by the authors to predict the energy content. While predicting the actual calorific value, the maximum error was 0.94% for the artificial neural network model and 5.20% by the statistical correlation. The network with eight neurons and an R2 = 0.96881 in the hidden layer results in a stable and optimum network. This study showed that the artificial neural network approach could successfully be used for energy content predictions from the municipal solid wastes in Northern Nigeria and other areas of similar waste stream and composition.

Variational Infinite Hidden Conditional Random Fields

NARCIS (Netherlands)

Bousmalis, Konstantinos; Zafeiriou, Stefanos; Morency, Louis-Philippe; Pantic, Maja; Ghahramani, Zoubin

2015-01-01

Hidden conditional random fields (HCRFs) are discriminative latent variable models which have been shown to successfully learn the hidden structure of a given classification problem. An Infinite hidden conditional random field is a hidden conditional random field with a countably infinite number of

Drift chamber tracking with neural networks

International Nuclear Information System (INIS)

Lindsey, C.S.; Denby, B.; Haggerty, H.

1992-10-01

We discuss drift chamber tracking with a commercial log VLSI neural network chip. Voltages proportional to the drift times in a 4-layer drift chamber were presented to the Intel ETANN chip. The network was trained to provide the intercept and slope of straight tracks traversing the chamber. The outputs were recorded and later compared off line to conventional track fits. Two types of network architectures were studied. Applications of neural network tracking to high energy physics detector triggers is discussed

Suitability assessment of artificial neural network to approximate surface subsidence due to rock mass drainage

Directory of Open Access Journals (Sweden)

Ryszard Hejmanowski

2015-01-01

Full Text Available Based on the previous studies conducted by the authors, a new approach was proposed, namely the tools of artificial intelligence. One of neural networks is a multilayer perceptron network (MLP, which has already found applications in many fields of science. Sequentially, a series of calculations was made for different MLP neural network configuration and the best of them was selected. Mean square error (MSE and the correlation coefficient R were adopted as the selection criterion for the optimal network. The obtained results were characterized with a considerable dispersion. With an increase in the amount of hidden neurons, the MSE of the network increased while the correlation coefficient R decreased. Similar conclusions were drawn for the network with a small number of hidden neurons. The analysis allowed to select a network composed of 24 neurons as the best one for the issue under question. The obtained final answers of artificial neural network were presented in a histogram as differences between the calculated and expected value.

Hidden charged dark matter

International Nuclear Information System (INIS)

Feng, Jonathan L.; Kaplinghat, Manoj; Tu, Huitzu; Yu, Hai-Bo

2009-01-01

Can dark matter be stabilized by charge conservation, just as the electron is in the standard model? We examine the possibility that dark matter is hidden, that is, neutral under all standard model gauge interactions, but charged under an exact (\\rm U)(1) gauge symmetry of the hidden sector. Such candidates are predicted in WIMPless models, supersymmetric models in which hidden dark matter has the desired thermal relic density for a wide range of masses. Hidden charged dark matter has many novel properties not shared by neutral dark matter: (1) bound state formation and Sommerfeld-enhanced annihilation after chemical freeze out may reduce its relic density, (2) similar effects greatly enhance dark matter annihilation in protohalos at redshifts of z ∼ 30, (3) Compton scattering off hidden photons delays kinetic decoupling, suppressing small scale structure, and (4) Rutherford scattering makes such dark matter self-interacting and collisional, potentially impacting properties of the Bullet Cluster and the observed morphology of galactic halos. We analyze all of these effects in a WIMPless model in which the hidden sector is a simplified version of the minimal supersymmetric standard model and the dark matter is a hidden sector stau. We find that charged hidden dark matter is viable and consistent with the correct relic density for reasonable model parameters and dark matter masses in the range 1 GeV ∼ X ∼< 10 TeV. At the same time, in the preferred range of parameters, this model predicts cores in the dark matter halos of small galaxies and other halo properties that may be within the reach of future observations. These models therefore provide a viable and well-motivated framework for collisional dark matter with Sommerfeld enhancement, with novel implications for astrophysics and dark matter searches

Localization of hidden Chua's attractors

International Nuclear Information System (INIS)

Leonov, G.A.; Kuznetsov, N.V.; Vagaitsev, V.I.

2011-01-01

The classical attractors of Lorenz, Rossler, Chua, Chen, and other widely-known attractors are those excited from unstable equilibria. From computational point of view this allows one to use numerical method, in which after transient process a trajectory, started from a point of unstable manifold in the neighborhood of equilibrium, reaches an attractor and identifies it. However there are attractors of another type: hidden attractors, a basin of attraction of which does not contain neighborhoods of equilibria. In the present Letter for localization of hidden attractors of Chua's circuit it is suggested to use a special analytical-numerical algorithm. -- Highlights: → There are hidden attractors: basin doesn't contain neighborhoods of equilibria. → Hidden attractors cannot be reached by trajectory from neighborhoods of equilibria. → We suggested special procedure for localization of hidden attractors. → We discovered hidden attractor in Chua's system, L. Chua in his work didn't expect this.

On-Line Condition Monitoring System for High Level Trip Water in Steam Boiler’s Drum

Directory of Open Access Journals (Sweden)

Ismail Alnaimi Firas B.

2014-07-01

Full Text Available This paper presents a monitoring technique using Artificial Neural Networks (ANN with four different training algorithms for high level water in steam boiler’s drum. Four Back-Propagations neural networks multidimensional minimization algorithms have been utilized. Real time data were recorded from power plant located in Malaysia. The developed relevant variables were selected based on a combination of theory, experience and execution phases of the model. The Root Mean Square (RMS Error has been used to compare the results of one and two hidden layer (1HL, (2HL ANN structures

Searching for hidden-charm baryonium signals in QCD sum rules

Energy Technology Data Exchange (ETDEWEB)

Chen, Hua-Xing; Zhou, Dan [Beihang University, School of Physics, Beijing Key Laboratory of Advanced Nuclear Materials and Physics, Beijing (China); Chen, Wei [University of Saskatchewan, Department of Physics and Engineering Physics, Saskatoon, SK (Canada); Liu, Xiang [Lanzhou University, School of Physical Science and Technology, Lanzhou (China); Lanzhou University, Research Center for Hadron and CSR Physics, Institute of Modern Physics of CAS, Lanzhou (China); Zhu, Shi-Lin [Peking University, School of Physics, State Key Laboratory of Nuclear Physics and Technology, Beijing (China); Collaborative Innovation Center of Quantum Matter, Beijing (China); Peking University, Center of High Energy Physics, Beijing (China)

2016-11-15

We give an explicit QCD sum rule investigation for hidden-charm baryonium states with the quark content u anti ud anti dc anti c, spin J = 0/1/2/3, and of both positive and negative parities. We systematically construct the relevant local hidden-charm baryonium interpolating currents, which can actually couple to various structures, including hidden-charm baryonium states, charmonium states plus two pions, and hidden-charm tetraquark states plus one pion, etc. We do not know which structure these currents couple to at the beginning, but after sum rule analyses we can obtain some information. We find some of them can couple to hidden-charm baryonium states, using which we evaluate the masses of the lowest-lying hidden-charm baryonium states with quantum numbers J{sup P} = 2{sup -}/3{sup -}/0{sup +}/1{sup +}/2{sup +} to be around 5.0 GeV. We suggest to search for hidden-charm baryonium states, especially the one of J = 3{sup -}, in the D-wave J/ψππ and P-wave J/ψρ and J/ψω channels in this energy region. (orig.)

Hidden Liquidity: Determinants and Impact

OpenAIRE

Gökhan Cebiroglu; Ulrich Horst

2012-01-01

We cross-sectionally analyze the presence of aggregated hidden depth and trade volume in the S&P 500 and identify its key determinants. We find that the spread is the main predictor for a stock’s hidden dimension, both in terms of traded and posted liquidity. Our findings moreover suggest that large hidden orders are associated with larger transaction costs, higher price impact and increased volatility. In particular, as large hidden orders fail to attract (latent) liquidity to the market, ...

Prediction of Daily Global Solar Radiation by Daily Temperatures and Artificial Neural Networks in Different Climates

Directory of Open Access Journals (Sweden)

S. I Saedi

2018-03-01

Full Text Available Introduction Global solar radiation is the sum of direct, diffuse, and reflected solar radiation. Weather forecasts, agricultural practices, and solar equipment development are three major fields that need proper information about solar radiation. Furthermore, sun in regarded as a huge source of renewable and clean energy which can be used in numerous applications to get rid of environmental impacts of non-renewable fossil fuels. Therefore, easy and fast estimation of daily global solar radiation would play an effective role is these affairs. Materials and Methods This study aimed at predicting the daily global solar radiation by means of artificial neural network (ANN method, based on easy-to-gain weather data i.e. daily mean, minimum and maximum temperatures. Having a variety of climates with long-term valid weather data, Washington State, located at the northwestern part of USA was chosen for this purpose. It has a total number of 19 weather stations to cover all the State climates. First, a station with the largest number of valid historical weather data (Lind was chosen to develop, validate, and test different ANN models. Three training algorithms i.e. Levenberg – Marquardt (LM, Scaled Conjugate Gradient (SCG, and Bayesian regularization (BR were tested in one and two hidden layer networks each with up to 20 neurons to derive six best architectures. R, RMSE, MAPE, and scatter plots were considered to evaluate each network in all steps. In order to investigate the generalizability of the best six models, they were tested in other Washington State weather stations. The most accurate and general models was evaluated in an Iran sample weather station which was chosen to be Mashhad. Results and Discussion The variation of MSE for the three training functions in one hidden layer models for Lind station indicated that SCG converged weights and biases in shorter time than LM, and LM did that faster than BR. It means that SCG provided the fastest

Analysis of Boiler Operational Variables Prior to Tube Leakage Fault by Artificial Intelligent System

Directory of Open Access Journals (Sweden)

Al-Kayiem Hussain H.

2014-07-01

Full Text Available Steam boilers are considered as a core of any steam power plant. Boilers are subjected to various types of trips leading to shut down of the entire plant. The tube leakage is the worse among the common boiler faults, where the shutdown period lasts for around four to five days. This paper describes the rules of the Artificial Intelligent Systems to diagnosis the boiler variables prior to tube leakage occurrence. An Intelligent system based on Artificial Neural Network was designed and coded in MATLAB environment. The ANN was trained and validated using real site data acquired from coal fired power plant in Malaysia. Ninety three boiler operational variables were identified for the present investigation based on the plant operator experience. Various neural networks topology combinations were investigated. The results showed that the NN with two hidden layers performed better than one hidden layer using Levenberg-Maquardt training algorithm. Moreover, it was noticed that hyperbolic tangent function for input and output nodes performed better than other activation function types.

Modeling and Prediction of Coal Ash Fusion Temperature based on BP Neural Network

Directory of Open Access Journals (Sweden)

Miao Suzhen

2016-01-01

Full Text Available Coal ash is the residual generated from combustion of coal. The ash fusion temperature (AFT of coal gives detail information on the suitability of a coal source for gasification procedures, and specifically to which extent ash agglomeration or clinkering is likely to occur within the gasifier. To investigate the contribution of oxides in coal ash to AFT, data of coal ash chemical compositions and Softening Temperature (ST in different regions of China were collected in this work and a BP neural network model was established by XD-APC PLATFORM. In the BP model, the inputs were the ash compositions and the output was the ST. In addition, the ash fusion temperature prediction model was obtained by industrial data and the model was generalized by different industrial data. Compared to empirical formulas, the BP neural network obtained better results. By different tests, the best result and the best configurations for the model were obtained: hidden layer nodes of the BP network was setted as three, the component contents (SiO2, Al2O3, Fe2O3, CaO, MgO were used as inputs and ST was used as output of the model.

Hidden gauge structure of supersymmetric free differential algebras

Energy Technology Data Exchange (ETDEWEB)

Andrianopoli, Laura [DISAT, Politecnico di Torino,Corso Duca degli Abruzzi 24, I-10129 Turin (Italy); INFN - Sezione di Torino,Torino (Italy); D’Auria, Riccardo [DISAT, Politecnico di Torino,Corso Duca degli Abruzzi 24, I-10129 Turin (Italy); Ravera, Lucrezia [DISAT, Politecnico di Torino,Corso Duca degli Abruzzi 24, I-10129 Turin (Italy); INFN - Sezione di Torino,Torino (Italy)

2016-08-16

The aim of this paper is to clarify the role of the nilpotent fermionic generator Q{sup ′} introduced in http://dx.doi.org/10.1016/0550-3213(82)90376-5 and appearing in the hidden supergroup underlying the free differential algebra (FDA) of D=11 supergravity. We give a physical explanation of its role by looking at the gauge properties of the theory. We find that its presence is necessary, in order that the extra 1-forms of the hidden supergroup give rise to the correct gauge transformations of the p-forms of the FDA. This interpretation is actually valid for any supergravity containing antisymmetric tensor fields, and any supersymmetric FDA can always be traded for a hidden Lie superalgebra containing extra fermionic nilpotent generators. As an interesting example we construct the hidden superalgebra associated with the FDA of N=2, D=7 supergravity. In this case we are able to parametrize the mutually non local 2- and 3-form B{sup (2)} and B{sup (3)} in terms of hidden 1-forms and find that supersymmetry and gauge invariance require in general the presence of two nilpotent fermionic generators in the hidden algebra. We propose that our approach, where all the invariances of the FDA are expressed as Lie derivatives of the p-forms in the hidden supergroup manifold, could be an appropriate framework to discuss theories defined in enlarged versions of superspace recently considered in the literature, such us double field theory and its generalizations.

New limits on hidden photons from past electron beam dumps

International Nuclear Information System (INIS)

Andreas, Sarah; Niebuhr, Carsten; Ringwald, Andreas

2012-09-01

Hidden sectors with light extra U(1) gauge bosons, so called hidden photons, have recently attracted some attention because they are a common feature of physics beyond the Standard Model like string theory and SUSY and additionally are phenomenologically of great interest regarding recent astrophysical observations. The hidden photon is already constrained by various laboratory experiments and presently searched for in running as well as upcoming experiments. We summarize the current status of limits on hidden photons from past electron beam dump experiments including two new limits from such experiments at KEK and Orsay that have so far not been considered. All our limits take into account the experimental acceptances obtained from Monte Carlo simulations.

New limits on hidden photons from past electron beam dumps

Energy Technology Data Exchange (ETDEWEB)

Andreas, Sarah; Niebuhr, Carsten; Ringwald, Andreas

2012-09-15

Hidden sectors with light extra U(1) gauge bosons, so called hidden photons, have recently attracted some attention because they are a common feature of physics beyond the Standard Model like string theory and SUSY and additionally are phenomenologically of great interest regarding recent astrophysical observations. The hidden photon is already constrained by various laboratory experiments and presently searched for in running as well as upcoming experiments. We summarize the current status of limits on hidden photons from past electron beam dump experiments including two new limits from such experiments at KEK and Orsay that have so far not been considered. All our limits take into account the experimental acceptances obtained from Monte Carlo simulations.

Classifying images using restricted Boltzmann machines and convolutional neural networks

Science.gov (United States)

Zhao, Zhijun; Xu, Tongde; Dai, Chenyu

2017-07-01

To improve the feature recognition ability of deep model transfer learning, we propose a hybrid deep transfer learning method for image classification based on restricted Boltzmann machines (RBM) and convolutional neural networks (CNNs). It integrates learning abilities of two models, which conducts subject classification by exacting structural higher-order statistics features of images. While the method transfers the trained convolutional neural networks to the target datasets, fully-connected layers can be replaced by restricted Boltzmann machine layers; then the restricted Boltzmann machine layers and Softmax classifier are retrained, and BP neural network can be used to fine-tuned the hybrid model. The restricted Boltzmann machine layers has not only fully integrated the whole feature maps, but also learns the statistical features of target datasets in the view of the biggest logarithmic likelihood, thus removing the effects caused by the content differences between datasets. The experimental results show that the proposed method has improved the accuracy of image classification, outperforming other methods on Pascal VOC2007 and Caltech101 datasets.

Multi-layer universal correction magnet

International Nuclear Information System (INIS)

Parzen, G.

1981-08-01

This paper presents an approach for constructing a universal correction magnet in which the return currents play an active role in determining the field. The return currents are not hidden by the iron shield. The coil is wound in many layers, instead of just one layer. Each layer has a particular symmetry, and generates a particular class of field multipoles such that the location of the return current for each independently excited current block is clear. Three layers may be sufficient in many cases. This approach is applied to the ISABELLE storage accelerator correction system

Optimization of a neural network model for signal-to-background prediction in gamma-ray spectrometry

International Nuclear Information System (INIS)

Dragovic, S.; Onjia, A. . E-mail address of corresponding author: sdragovic@inep.co.yu; Dragovic, S.)

2005-01-01

The artificial neural network (ANN) model was optimized for the prediction of signal-to-background (SBR) ratio as a function of the measurement time in gamma-ray spectrometry. The network parameters: learning rate (α), momentum (μ), number of epochs (E) and number of nodes in hidden layer (N) were optimized simultaneously employing variable-size simplex method. The most accurate model with the root mean square (RMS) error of 0.073 was obtained using ANN with online backpropagation randomized (OBPR) algorithm with α = 0.27, μ 0.36, E = 14800 and N = 9. Most of the predicted and experimental SBR values for the eight radionuclides ( 226 Ra, 214 Bi, 235 U, 40 K, 232 Th, 134 Cs, 137 Cs and 7 Be), studied in this work, reasonably agreed to within 15 %, which was satisfactory accuracy. (author)

Detection of Coal Fires: A Case Study Conducted on Indian Coal Seams Using Neural Network and Particle Swarm Optimization

Science.gov (United States)

Singh, B. B.

2016-12-01

India produces majority of its electricity from coal but a huge quantity of coal burns every day due to coal fires and also poses a threat to the environment as severe pollutants. In the present study we had demonstrated the usage of Neural Network based approach with an integrated Particle Swarm Optimization (PSO) inversion technique. The Self Potential (SP) data set is used for the early detection of coal fires. The study was conducted over the East Basuria colliery, Jharia Coal Field, Jharkhand, India. The causative source was modelled as an inclined sheet like anomaly and the synthetic data was generated. Neural Network scheme consists of an input layer, hidden layers and an output layer. The input layer corresponds to the SP data and the output layer is the estimated depth of the coal fire. A synthetic dataset was modelled with some of the known parameters such as depth, conductivity, inclination angle, half width etc. associated with causative body and gives a very low misfit error of 0.0032%. Therefore, the method was found accurate in predicting the depth of the source body. The technique was applied to the real data set and the model was trained until a very good correlation of determination `R2' value of 0.98 is obtained. The depth of the source body was found to be 12.34m with a misfit error percentage of 0.242%. The inversion results were compared with the lithologs obtained from a nearby well which corresponds to the L3 coal seam. The depth of the coal fire had exactly matched with the half width of the anomaly which suggests that the fire is widely spread. The inclination angle of the anomaly was 135.510 which resembles the development of the geometrically complex fracture planes. These fractures may be developed due to anisotropic weakness of the ground which acts as passage for the air. As a result coal fires spreads along these fracture planes. The results obtained from the Neural Network was compared with PSO inversion results and were found in

SU-F-E-09: Respiratory Signal Prediction Based On Multi-Layer Perceptron Neural Network Using Adjustable Training Samples

Energy Technology Data Exchange (ETDEWEB)

Sun, W; Jiang, M; Yin, F [Duke University Medical Center, Durham, NC (United States)

2016-06-15

Purpose: Dynamic tracking of moving organs, such as lung and liver tumors, under radiation therapy requires prediction of organ motions prior to delivery. The shift of moving organ may change a lot due to huge transform of respiration at different periods. This study aims to reduce the influence of that changes using adjustable training signals and multi-layer perceptron neural network (ASMLP). Methods: Respiratory signals obtained using a Real-time Position Management(RPM) device were used for this study. The ASMLP uses two multi-layer perceptron neural networks(MLPs) to infer respiration position alternately and the training sample will be updated with time. Firstly, a Savitzky-Golay finite impulse response smoothing filter was established to smooth the respiratory signal. Secondly, two same MLPs were developed to estimate respiratory position from its previous positions separately. Weights and thresholds were updated to minimize network errors according to Leverberg-Marquart optimization algorithm through backward propagation method. Finally, MLP 1 was used to predict 120∼150s respiration position using 0∼120s training signals. At the same time, MLP 2 was trained using 30∼150s training signals. Then MLP is used to predict 150∼180s training signals according to 30∼150s training signals. The respiration position is predicted as this way until it was finished. Results: In this experiment, the two methods were used to predict 2.5 minute respiratory signals. For predicting 1s ahead of response time, correlation coefficient was improved from 0.8250(MLP method) to 0.8856(ASMLP method). Besides, a 30% improvement of mean absolute error between MLP(0.1798 on average) and ASMLP(0.1267 on average) was achieved. For predicting 2s ahead of response time, correlation coefficient was improved from 0.61415 to 0.7098.Mean absolute error of MLP method(0.3111 on average) was reduced by 35% using ASMLP method(0.2020 on average). Conclusion: The preliminary results

Generalized Projective Synchronization between Two Different Neural Networks with Mixed Time Delays

Directory of Open Access Journals (Sweden)

Xuefei Wu

2012-01-01

Full Text Available The generalized projective synchronization (GPS between two different neural networks with nonlinear coupling and mixed time delays is considered. Several kinds of nonlinear feedback controllers are designed to achieve GPS between two different such neural networks. Some results for GPS of these neural networks are proved theoretically by using the Lyapunov stability theory and the LaSalle invariance principle. Moreover, by comparison, we determine an optimal nonlinear controller from several ones and provide an adaptive update law for it. Computer simulations are provided to show the effectiveness and feasibility of the proposed methods.

NETS - A NEURAL NETWORK DEVELOPMENT TOOL, VERSION 3.0 (MACHINE INDEPENDENT VERSION)

Science.gov (United States)

Baffes, P. T.

1994-01-01

NETS, A Tool for the Development and Evaluation of Neural Networks, provides a simulation of Neural Network algorithms plus an environment for developing such algorithms. Neural Networks are a class of systems modeled after the human brain. Artificial Neural Networks are formed from hundreds or thousands of simulated neurons, connected to each other in a manner similar to brain neurons. Problems which involve pattern matching readily fit the class of problems which NETS is designed to solve. NETS uses the back propagation learning method for all of the networks which it creates. The nodes of a network are usually grouped together into clumps called layers. Generally, a network will have an input layer through which the various environment stimuli are presented to the network, and an output layer for determining the network's response. The number of nodes in these two layers is usually tied to some features of the problem being solved. Other layers, which form intermediate stops between the input and output layers, are called hidden layers. NETS allows the user to customize the patterns of connections between layers of a network. NETS also provides features for saving the weight values of a network during the learning process, which allows for more precise control over the learning process. NETS is an interpreter. Its method of execution is the familiar "read-evaluate-print" loop found in interpreted languages such as BASIC and LISP. The user is presented with a prompt which is the simulator's way of asking for input. After a command is issued, NETS will attempt to evaluate the command, which may produce more prompts requesting specific information or an error if the command is not understood. The typical process involved when using NETS consists of translating the problem into a format which uses input/output pairs, designing a network configuration for the problem, and finally training the network with input/output pairs until an acceptable error is reached. NETS

Sensorless control for permanent magnet synchronous motor using a neural network based adaptive estimator

Science.gov (United States)

Kwon, Chung-Jin; Kim, Sung-Joong; Han, Woo-Young; Min, Won-Kyoung

2005-12-01

The rotor position and speed estimation of permanent-magnet synchronous motor(PMSM) was dealt with. By measuring the phase voltages and currents of the PMSM drive, two diagonally recurrent neural network(DRNN) based observers, a neural current observer and a neural velocity observer were developed. DRNN which has self-feedback of the hidden neurons ensures that the outputs of DRNN contain the whole past information of the system even if the inputs of DRNN are only the present states and inputs of the system. Thus the structure of DRNN may be simpler than that of feedforward and fully recurrent neural networks. If the backpropagation method was used for the training of the DRNN the problem of slow convergence arise. In order to reduce this problem, recursive prediction error(RPE) based learning method for the DRNN was presented. The simulation results show that the proposed approach gives a good estimation of rotor speed and position, and RPE based training has requires a shorter computation time compared to backpropagation based training.

Classification of Company Performance using Weighted Probabilistic Neural Network

Science.gov (United States)

Yasin, Hasbi; Waridi Basyiruddin Arifin, Adi; Warsito, Budi

2018-05-01

Classification of company performance can be judged by looking at its financial status, whether good or bad state. Classification of company performance can be achieved by some approach, either parametric or non-parametric. Neural Network is one of non-parametric methods. One of Artificial Neural Network (ANN) models is Probabilistic Neural Network (PNN). PNN consists of four layers, i.e. input layer, pattern layer, addition layer, and output layer. The distance function used is the euclidean distance and each class share the same values as their weights. In this study used PNN that has been modified on the weighting process between the pattern layer and the addition layer by involving the calculation of the mahalanobis distance. This model is called the Weighted Probabilistic Neural Network (WPNN). The results show that the company's performance modeling with the WPNN model has a very high accuracy that reaches 100%.

Neural Networks for Non-linear Control

DEFF Research Database (Denmark)

Sørensen, O.

1994-01-01

This paper describes how a neural network, structured as a Multi Layer Perceptron, is trained to predict, simulate and control a non-linear process.......This paper describes how a neural network, structured as a Multi Layer Perceptron, is trained to predict, simulate and control a non-linear process....

Using Artificial Neural Networks to Model the Surface Roughness of Massive Wooden Edge-Glued Panels Made of Scotch Pine (Pinus sylvestris L. in a Machining Process with Computer Numerical Control

Directory of Open Access Journals (Sweden)

Sait Dundar Sofuoglu

2015-08-01

Full Text Available An artificial neural network (ANN approach was employed for the prediction and control of surface roughness (Ra and Rz in a computer numerical control (CNC machine. Experiments were performed on a CNC machine to obtain data used for the training and testing of an ANN. Experimental studies were conducted, and a model based on the experimental results was set up. Five machining parameters (cutter type, tool clearance strategy, spindle speed, feed rate, and depth of cut were used. One hidden layer was used for all models, while there were five neurons in the hidden layer of the Ra and Rz models. The RMSE values were calculated as 1.05 and 3.70. The mean absolute percentage error (MAPE values were calculated as 20.18 and 15.14, which can be considered as a good prediction. The results of the ANN approach were compared with the measured values. It was shown that the ANN prediction model obtained is a useful and effective tool for modeling the Ra and Rz of wood. The results of the present research can be applied in the wood machining industry to reduce energy, time, and cost.

Prediction of hearing loss among the noise-exposed workers in a steel factory using artificial intelligence approach.

Science.gov (United States)

Aliabadi, Mohsen; Farhadian, Maryam; Darvishi, Ebrahim

2015-08-01

Prediction of hearing loss in noisy workplaces is considered to be an important aspect of hearing conservation program. Artificial intelligence, as a new approach, can be used to predict the complex phenomenon such as hearing loss. Using artificial neural networks, this study aims to present an empirical model for the prediction of the hearing loss threshold among noise-exposed workers. Two hundred and ten workers employed in a steel factory were chosen, and their occupational exposure histories were collected. To determine the hearing loss threshold, the audiometric test was carried out using a calibrated audiometer. The personal noise exposure was also measured using a noise dosimeter in the workstations of workers. Finally, data obtained five variables, which can influence the hearing loss, were used for the development of the prediction model. Multilayer feed-forward neural networks with different structures were developed using MATLAB software. Neural network structures had one hidden layer with the number of neurons being approximately between 5 and 15 neurons. The best developed neural networks with one hidden layer and ten neurons could accurately predict the hearing loss threshold with RMSE = 2.6 dB and R(2) = 0.89. The results also confirmed that neural networks could provide more accurate predictions than multiple regressions. Since occupational hearing loss is frequently non-curable, results of accurate prediction can be used by occupational health experts to modify and improve noise exposure conditions.

Artificial neural network for suppression of banding artifacts in balanced steady-state free precession MRI.

Science.gov (United States)

Kim, Ki Hwan; Park, Sung-Hong

2017-04-01

The balanced steady-state free precession (bSSFP) MR sequence is frequently used in clinics, but is sensitive to off-resonance effects, which can cause banding artifacts. Often multiple bSSFP datasets are acquired at different phase cycling (PC) angles and then combined in a special way for banding artifact suppression. Many strategies of combining the datasets have been suggested for banding artifact suppression, but there are still limitations in their performance, especially when the number of phase-cycled bSSFP datasets is small. The purpose of this study is to develop a learning-based model to combine the multiple phase-cycled bSSFP datasets for better banding artifact suppression. Multilayer perceptron (MLP) is a feedforward artificial neural network consisting of three layers of input, hidden, and output layers. MLP models were trained by input bSSFP datasets acquired from human brain and knee at 3T, which were separately performed for two and four PC angles. Banding-free bSSFP images were generated by maximum-intensity projection (MIP) of 8 or 12 phase-cycled datasets and were used as targets for training the output layer. The trained MLP models were applied to another brain and knee datasets acquired with different scan parameters and also to multiple phase-cycled bSSFP functional MRI datasets acquired on rat brain at 9.4T, in comparison with the conventional MIP method. Simulations were also performed to validate the MLP approach. Both the simulations and human experiments demonstrated that MLP suppressed banding artifacts significantly, superior to MIP in both banding artifact suppression and SNR efficiency. MLP demonstrated superior performance over MIP for the 9.4T fMRI data as well, which was not used for training the models, while visually preserving the fMRI maps very well. Artificial neural network is a promising technique for combining multiple phase-cycled bSSFP datasets for banding artifact suppression. Copyright © 2016 Elsevier Inc. All

Multi-layer network utilizing rewarded spike time dependent plasticity to learn a foraging task.

Directory of Open Access Journals (Sweden)

Pavel Sanda

2017-09-01

Full Text Available Neural networks with a single plastic layer employing reward modulated spike time dependent plasticity (STDP are capable of learning simple foraging tasks. Here we demonstrate advanced pattern discrimination and continuous learning in a network of spiking neurons with multiple plastic layers. The network utilized both reward modulated and non-reward modulated STDP and implemented multiple mechanisms for homeostatic regulation of synaptic efficacy, including heterosynaptic plasticity, gain control, output balancing, activity normalization of rewarded STDP and hard limits on synaptic strength. We found that addition of a hidden layer of neurons employing non-rewarded STDP created neurons that responded to the specific combinations of inputs and thus performed basic classification of the input patterns. When combined with a following layer of neurons implementing rewarded STDP, the network was able to learn, despite the absence of labeled training data, discrimination between rewarding patterns and the patterns designated as punishing. Synaptic noise allowed for trial-and-error learning that helped to identify the goal-oriented strategies which were effective in task solving. The study predicts a critical set of properties of the spiking neuronal network with STDP that was sufficient to solve a complex foraging task involving pattern classification and decision making.

Development of neural network model of the multiparametric ...

African Journals Online (AJOL)