Search results for: Neural function.

Authors: Radouane Iqdour, Abdelouhab Zeroual

Abstract:

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 performed, and the obtained results show that the neural networks are more efficient and gave the best results.

Keywords: Daily solar radiation, Prediction, MLP neural networks, linear model

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1339

Authors: S. Mousavian, D. Ashouri, F. Mousavian, V. Nikkhah Rashidabad, N. Ghazinia

Abstract:

PH, temperature and time of extraction of each stage,  agitation speed and delay time between stages effect on efficiency of  zinc extraction from concentrate. In this research, efficiency of zinc  extraction was predicted as a function of mentioned variable by  artificial neural networks (ANN). ANN with different layer was  employed and the result show that the networks with 8 neurons in  hidden layer has good agreement with experimental data.

 

Keywords: Zinc extraction, Efficiency, Neural networks, Operating condition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1590

Authors: Sinarwati Mohamad Suhaili, Naomie Salim, Mohamad Nazim Jambli

Abstract:

Sequence-to-sequence (seq2seq) models augmented with attention mechanisms are increasingly important in automated customer service. These models, adept at recognizing complex relationships between input and output sequences, are essential for optimizing chatbot responses. Central to these mechanisms are neural attention weights that determine the model’s focus during sequence generation. Despite their widespread use, there remains a gap in the comparative analysis of different attention weighting functions within seq2seq models, particularly in the context of chatbots utilizing the Customer Support Twitter (CST) dataset. This study addresses this gap by evaluating four distinct attention-scoring functions—dot, multiplicative/general, additive, and an extended multiplicative function with a tanh activation parameter — in neural generative seq2seq models. Using the CST dataset, these models were trained and evaluated over 10 epochs with the AdamW optimizer. Evaluation criteria included validation loss and BLEU scores implemented under both greedy and beam search strategies with a beam size of k = 3. Results indicate that the model with the tanh-augmented multiplicative function significantly outperforms its counterparts, achieving the lowest validation loss (1.136484) and the highest BLEU scores (0.438926 under greedy search, 0.443000 under beam search, k = 3). These findings emphasize the crucial influence of selecting an appropriate attention-scoring function to enhance the performance of seq2seq models for chatbots, particularly highlighting the model integrating tanh activation as a promising approach to improving chatbot quality in customer support contexts.

Keywords: Attention weight, chatbot, encoder-decoder, neural generative attention, score function, sequence-to-sequence.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 102

Authors: Isao Taguchi, Yasuo Sugai

Abstract:

This paper proposes an efficient learning method for the layered neural networks based on the selection of training data and input characteristics of an output layer unit. Comparing to recent neural networks; pulse neural networks, quantum neuro computation, etc, the multilayer network is widely used due to its simple structure. When learning objects are complicated, the problems, such as unsuccessful learning or a significant time required in learning, remain unsolved. Focusing on the input data during the learning stage, we undertook an experiment to identify the data that makes large errors and interferes with the learning process. Our method devides the learning process into several stages. In general, input characteristics to an output layer unit show oscillation during learning process for complicated problems. The multi-stage learning method proposes by the authors for the function approximation problems of classifying learning data in a phased manner, focusing on their learnabilities prior to learning in the multi layered neural network, and demonstrates validity of the multi-stage learning method. Specifically, this paper verifies by computer experiments that both of learning accuracy and learning time are improved of the BP method as a learning rule of the multi-stage learning method. In learning, oscillatory phenomena of a learning curve serve an important role in learning performance. The authors also discuss the occurrence mechanisms of oscillatory phenomena in learning. Furthermore, the authors discuss the reasons that errors of some data remain large value even after learning, observing behaviors during learning.

Keywords: data selection, function approximation problem, multistage leaning, neural network, voluntary oscillation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1432

Authors: Zixin Liu, Jianjun Jiao Wanping Bai

Abstract:

In this paper, the issue of pth moment exponential stability of stochastic recurrent neural network with distributed time delays is investigated. By using the method of variation parameters, inequality techniques, and stochastic analysis, some sufficient conditions ensuring pth moment exponential stability are obtained. The method used in this paper does not resort to any Lyapunov function, and the results derived in this paper generalize some earlier criteria reported in the literature. One numerical example is given to illustrate the main results.

Keywords: Stochastic recurrent neural networks, pth moment exponential stability, distributed time delays.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1262

Authors: S.I Sulaiman, T.K Abdul Rahman, I. Musirin, S. Shaari

Abstract:

This paper presents performance analysis of the Evolutionary Programming-Artificial Neural Network (EPANN) based technique to optimize the architecture and training parameters of a one-hidden layer feedforward ANN model for the prediction of energy output from a grid connected photovoltaic system. The ANN utilizes solar radiation and ambient temperature as its inputs while the output is the total watt-hour energy produced from the grid-connected PV system. EP is used to optimize the regression performance of the ANN model by determining the optimum values for the number of nodes in the hidden layer as well as the optimal momentum rate and learning rate for the training. The EPANN model is tested using two types of transfer function for the hidden layer, namely the tangent sigmoid and logarithmic sigmoid. The best transfer function, neural topology and learning parameters were selected based on the highest regression performance obtained during the ANN training and testing process. It is observed that the best transfer function configuration for the prediction model is [logarithmic sigmoid, purely linear].

Keywords: Artificial neural network (ANN), Correlation coefficient (R), Evolutionary programming-ANN (EPANN), Photovoltaic (PV), logarithmic sigmoid and tangent sigmoid.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1903

Authors: Farhad Asadi, S. Hossein Sadati

Abstract:

This paper presents a prediction performance of feedforward Multilayer Perceptron (MLP) and Echo State Networks (ESN) trained with extended Kalman filter. Feedforward neural networks and ESN are powerful neural networks which can track and predict nonlinear signals. However, their tracking performance depends on the specific signals or data sets, having the risk of instability accompanied by large error. In this study we explore this process by applying different network size and leaking rate for prediction of nonlinear or chaotic signals in MLP neural networks. Major problems of ESN training such as the problem of initialization of the network and improvement in the prediction performance are tackled. The influence of coefficient of activation function in the hidden layer and other key parameters are investigated by simulation results. Extended Kalman filter is employed in order to improve the sequential and regulation learning rate of the feedforward neural networks. This training approach has vital features in the training of the network when signals have chaotic or non-stationary sequential pattern. Minimization of the variance in each step of the computation and hence smoothing of tracking were obtained by examining the results, indicating satisfactory tracking characteristics for certain conditions. In addition, simulation results confirmed satisfactory performance of both of the two neural networks with modified parameterization in tracking of the nonlinear signals.

Keywords: Feedforward neural networks, nonlinear signal prediction, echo state neural networks approach, leaking rates, capacity of neural networks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 763

Authors: Omid Borazjani, Mahmoud Roosta, Khodakhast Isapour, Ali Reza Rajabi

Abstract:

This paper presents an application of Artificial Neural Network (ANN) algorithm for improving power system voltage stability. The training data is obtained by solving several normal and abnormal conditions using the Linear Programming technique. The selected objective function gives minimum deviation of the reactive power control variables, which leads to the maximization of minimum Eigen value of load flow Jacobian. The considered reactive power control variables are switchable VAR compensators, OLTC transformers and excitation of generators. The method has been implemented on a modified IEEE 30-bus test system. The results obtain from the test clearly show that the trained neural network is capable of improving the voltage stability in power system with a high level of precision and speed.

Keywords: Artificial Neural Network (ANN), Load Flow, Voltage Stability, Power Systems.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1978

Authors: Mokrane Selma

Abstract:

The shift towards decision making (DM) based on artificial intelligence (AI) techniques will change the way in which consumer markets and our societies function. Through AI, predictive analytics is being used by businesses to identify these patterns and major trends with the objective to improve the DM and influence future business outcomes. This paper proposes an Artificial Neural Network (ANN) approach to predict the success of telemarketing calls for selling bank long-term deposits. To validate the proposed model, we uses the bank marketing data of 41188 phone calls. The ANN attains 98.93% of accuracy which outperforms other conventional classifiers and confirms that it is credible and valuable approach for telemarketing campaign managers.

Keywords: Bank telemarketing, prediction, decision making, artificial intelligence, artificial neural network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3158

Authors: K.-J. You, H.J. Lee, Y. Lang, C. Im, C.S. Koh, H.-C. Shin

Abstract:

This paper presents a novel method for inferring the odor based on neural activities observed from rats- main olfactory bulbs. Multi-channel extra-cellular single unit recordings were done by micro-wire electrodes (tungsten, 50μm, 32 channels) implanted in the mitral/tufted cell layers of the main olfactory bulb of anesthetized rats to obtain neural responses to various odors. Neural response as a key feature was measured by substraction of neural firing rate before stimulus from after. For odor inference, we have developed a decoding method based on the maximum likelihood (ML) estimation. The results have shown that the average decoding accuracy is about 100.0%, 96.0%, 84.0%, and 100.0% with four rats, respectively. This work has profound implications for a novel brain-machine interface system for odor inference.

Keywords: biomedical signal processing, neural engineering, olfactory, neural decoding, BMI

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1620

Authors: Rui Antunes, Fernando V. Coito

Abstract:

This article presents the development of a neural network cognitive model for the classification and detection of different frequency signals. The basic structure of the implemented neural network was inspired on the perception process that humans generally make in order to visually distinguish between high and low frequency signals. It is based on the dynamic neural network concept, with delays. A special two-layer feedforward neural net structure was successfully implemented, trained and validated, to achieve minimum target error. Training confirmed that this neural net structure descents and converges to a human perception classification solution, even when far away from the target.

Keywords: Neural Networks, Signal Classification, Adaptative Filters, Cognitive Neuroscience

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1675

Authors: Shu Lü, Shouming Zhong, Zixin Liu

Abstract:

In this paper, the robust exponential stability problem of discrete-time uncertain stochastic neural networks with timevarying delays is investigated. By introducing a new augmented Lyapunov function, some delay-dependent stable results are obtained in terms of linear matrix inequality (LMI) technique. Compared with some existing results in the literature, the conservatism of the new criteria is reduced notably. Three numerical examples are provided to demonstrate the less conservatism and effectiveness of the proposed method.

Keywords: Robust exponential stability, delay-dependent stability, discrete-time neural networks, stochastic, time-varying delays.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1439

Authors: B. Chandra, P. Paul Varghese

Abstract:

Crypto System Identification is one of the challenging tasks in Crypt analysis. The paper discusses the possibility of employing Neural Networks for identification of Cipher Systems from cipher texts. Cascade Correlation Neural Network and Back Propagation Network have been employed for identification of Cipher Systems. Very large collection of cipher texts were generated using a Block Cipher (Enhanced RC6) and a Stream Cipher (SEAL). Promising results were obtained in terms of accuracy using both the Neural Network models but it was observed that the Cascade Correlation Neural Network Model performed better compared to Back Propagation Network.

Keywords: Back Propagation Neural Networks, CascadeCorrelation Neural Network, Crypto systems, Block Cipher, StreamCipher.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2448

Authors: Thomas Behan, Zaiyi Liao, Lian Zhao, Chunting Yang

Abstract:

In this paper, low end Digital Signal Processors (DSPs) are applied to accelerate integer neural networks. The use of DSPs to accelerate neural networks has been a topic of study for some time, and has demonstrated significant performance improvements. Recently, work has been done on integer only neural networks, which greatly reduces hardware requirements, and thus allows for cheaper hardware implementation. DSPs with Arithmetic Logic Units (ALUs) that support floating or fixed point arithmetic are generally more expensive than their integer only counterparts due to increased circuit complexity. However if the need for floating or fixed point math operation can be removed, then simpler, lower cost DSPs can be used. To achieve this, an integer only neural network is created in this paper, which is then accelerated by using DSP instructions to improve performance.

Keywords: Digital Signal Processor (DSP), Integer Neural Network(INN), Low Cost Neural Network, Integer Neural Network DSPImplementation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1803

Authors: N. M. Nawi, M. R. Ransing, R. S. Ransing

Abstract:

The conjugate gradient optimization algorithm usually used for nonlinear least squares is presented and is combined with the modified back propagation algorithm yielding a new fast training multilayer perceptron (MLP) algorithm (CGFR/AG). The approaches presented in the paper consist of three steps: (1) Modification on standard back propagation algorithm by introducing gain variation term of the activation function, (2) Calculating the gradient descent on error with respect to the weights and gains values and (3) the determination of the new search direction by exploiting the information calculated by gradient descent in step (2) as well as the previous search direction. The proposed method improved the training efficiency of back propagation algorithm by adaptively modifying the initial search direction. Performance of the proposed method is demonstrated by comparing to the conjugate gradient algorithm from neural network toolbox for the chosen benchmark. The results show that the number of iterations required by the proposed method to converge is less than 20% of what is required by the standard conjugate gradient and neural network toolbox algorithm.

Keywords: Back-propagation, activation function, conjugategradient, search direction, gain variation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2842

Authors: Siddhartha Bhattacharyya, Paramartha Dutta, Ujjwal Maulik, Prashanta Kumar Nandi

Abstract:

The paper describes a self supervised parallel self organizing neural network (PSONN) architecture for true color image segmentation. The proposed architecture is a parallel extension of the standard single self organizing neural network architecture (SONN) and comprises an input (source) layer of image information, three single self organizing neural network architectures for segmentation of the different primary color components in a color image scene and one final output (sink) layer for fusion of the segmented color component images. Responses to the different shades of color components are induced in each of the three single network architectures (meant for component level processing) by applying a multilevel version of the characteristic activation function, which maps the input color information into different shades of color components, thereby yielding a processed component color image segmented on the basis of the different shades of component colors. The number of target classes in the segmented image corresponds to the number of levels in the multilevel activation function. Since the multilevel version of the activation function exhibits several subnormal responses to the input color image scene information, the system errors of the three component network architectures are computed from some subnormal linear index of fuzziness of the component color image scenes at the individual level. Several multilevel activation functions are employed for segmentation of the input color image scene using the proposed network architecture. Results of the application of the multilevel activation functions to the PSONN architecture are reported on three real life true color images. The results are substantiated empirically with the correlation coefficients between the segmented images and the original images.

Keywords: Colour image segmentation, fuzzy set theory, multi-level activation functions, parallel self-organizing neural network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2025

Authors: Tao Peng, Jing Zhao, Yanqing Xu, Jing Cai

Abstract:

Due to missing/ambiguous boundaries between the prostate and neighboring structures, the presence of shadow artifacts, as well as the large variability in prostate shapes, ultrasound prostate segmentation is challenging. To handle these issues, this paper develops a hybrid method for ultrasound prostate segmentation by combining an optimized closed principal curve-based method and the evolutionary neural network; the former can fit curves with great curvature and generate a contour composed of line segments connected by sorted vertices, and the latter is used to express an appropriate map function (represented by parameters of evolutionary neural network) for generating the smooth prostate contour to match the ground truth contour. Both qualitative and quantitative experimental results showed that our proposed method obtains accurate and robust performances.

Keywords: Ultrasound prostate segmentation, optimized closed polygonal segment method, evolutionary neural network, smooth mathematical model.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 464

Authors: Hazem M. El-Bakry, Qiangfu Zhao

Abstract:

In this paper, an approach to reduce the computation steps required by fast neural networksfor the searching process is presented. The principle ofdivide and conquer strategy is applied through imagedecomposition. Each image is divided into small in sizesub-images and then each one is tested separately usinga fast neural network. The operation of fast neuralnetworks based on applying cross correlation in thefrequency domain between the input image and theweights of the hidden neurons. Compared toconventional and fast neural networks, experimentalresults show that a speed up ratio is achieved whenapplying this technique to locate human facesautomatically in cluttered scenes. Furthermore, fasterface detection is obtained by using parallel processingtechniques to test the resulting sub-images at the sametime using the same number of fast neural networks. Incontrast to using only fast neural networks, the speed upratio is increased with the size of the input image whenusing fast neural networks and image decomposition.

Keywords: Fast Neural Networks, 2D-FFT, CrossCorrelation, Image decomposition, Parallel Processing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2186

Authors: S. Janpong, K-L. Areerak, K-N. Areerak

Abstract:

This paper aims to present the reviews of the application of neural network in shunt active power filter (SAPF). From the review, three out of four components of SAPF structure, which are harmonic detection component, compensating current control, and DC bus voltage control, have been adopted some of neural network architecture as part of its component or even substitution. The objectives of most papers in using neural network in SAPF are to increase the efficiency, stability, accuracy, robustness, tracking ability of the systems of each component. Moreover, minimizing unneeded signal due to the distortion is the ultimate goal in applying neural network to the SAPF. The most famous architecture of neural network in SAPF applications are ADALINE and Backpropagation (BP).

Keywords: Active power filter, neural network, harmonic distortion, harmonic detection and compensation, non-linear load.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3069

Authors: Siddhartha Bhattacharyya, Paramartha Dutta, Ujjwal Maulik, Prashanta Kumar Nandi

Abstract:

A multilayer self organizing neural neural network (MLSONN) architecture for binary object extraction, guided by a beta activation function and characterized by backpropagation of errors estimated from the linear indices of fuzziness of the network output states, is discussed. Since the MLSONN architecture is designed to operate in a single point fixed/uniform thresholding scenario, it does not take into cognizance the heterogeneity of image information in the extraction process. The performance of the MLSONN architecture with representative values of the threshold parameters of the beta activation function employed is also studied. A three layer bidirectional self organizing neural network (BDSONN) architecture comprising fully connected neurons, for the extraction of objects from a noisy background and capable of incorporating the underlying image context heterogeneity through variable and adaptive thresholding, is proposed in this article. The input layer of the network architecture represents the fuzzy membership information of the image scene to be extracted. The second layer (the intermediate layer) and the final layer (the output layer) of the network architecture deal with the self supervised object extraction task by bi-directional propagation of the network states. Each layer except the output layer is connected to the next layer following a neighborhood based topology. The output layer neurons are in turn, connected to the intermediate layer following similar topology, thus forming a counter-propagating architecture with the intermediate layer. The novelty of the proposed architecture is that the assignment/updating of the inter-layer connection weights are done using the relative fuzzy membership values at the constituent neurons in the different network layers. Another interesting feature of the network lies in the fact that the processing capabilities of the intermediate and the output layer neurons are guided by a beta activation function, which uses image context sensitive adaptive thresholding arising out of the fuzzy cardinality estimates of the different network neighborhood fuzzy subsets, rather than resorting to fixed and single point thresholding. An application of the proposed architecture for object extraction is demonstrated using a synthetic and a real life image. The extraction efficiency of the proposed network architecture is evaluated by a proposed system transfer index characteristic of the network.

Keywords: Beta activation function, fuzzy cardinality, multilayer self organizing neural network, object extraction,

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1570

Authors: Ahmed Abdolkhalig

Abstract:

This paper examines many mathematical methods for molding the hourly price forward curve (HPFC); the model will be constructed by numerous regression methods, like polynomial regression, radial basic function neural networks & a furrier series. Examination the models goodness of fit will be done by means of statistical & graphical tools. The criteria for choosing the model will depend on minimize the Root Mean Squared Error (RMSE), using the correlation analysis approach for the regression analysis the optimal model will be distinct, which are robust against model misspecification. Learning & supervision technique employed to determine the form of the optimal parameters corresponding to each measure of overall loss. By using all the numerical methods that mentioned previously; the explicit expressions for the optimal model derived and the optimal designs will be implemented.

Keywords: Forward curve, furrier series, regression, radial basic function neural networks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4234

Authors: Hazem M. El-Bakry, Qiangfu Zhao

Abstract:

Here, a new idea to speed up the operation of complex valued time delay neural networks is presented. The whole data are collected together in a long vector and then tested as a one input pattern. The proposed fast complex valued time delay neural networks uses cross correlation in the frequency domain between the tested data and the input weights of neural networks. It is proved mathematically that the number of computation steps required for the presented fast complex valued time delay neural networks is less than that needed by classical time delay neural networks. Simulation results using MATLAB confirm the theoretical computations.

Keywords: Fast Complex Valued Time Delay Neural Networks, Cross Correlation, Frequency Domain

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1828

Authors: Kyu Jin Kim, Byung Kwan Oh, Hyo Seon Park

Abstract:

The seismic responses-based structural health monitoring system has been performed to prevent seismic damage. Structural seismic damage of building is caused by the instantaneous stress concentration which is related with dynamic characteristic of earthquake. Meanwhile, seismic response analysis to estimate the dynamic responses of building demands significantly high computational cost. To prevent the failure of structural members from the characteristic of the earthquake and the significantly high computational cost for seismic response analysis, this paper presents an artificial neural network (ANN) based prediction model for dynamic responses of building considering specific time length. Through the measured dynamic responses, input and output node of the ANN are formed by the length of specific time, and adopted for the training. In the model, evolutionary radial basis function neural network (ERBFNN), that radial basis function network (RBFN) is integrated with evolutionary optimization algorithm to find variables in RBF, is implemented. The effectiveness of the proposed model is verified through an analytical study applying responses from dynamic analysis for multi-degree of freedom system to training data in ERBFNN.

Keywords: Structural health monitoring, dynamic response, artificial neural network, radial basis function network, genetic algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 425

Authors: Khalid Salhi, El Miloud Jaara, Mohammed Talibi Alaoui

Abstract:

In this paper, we present a neural approach for unsupervised natural color-texture image segmentation, which is based on both Kohonen maps and mathematical morphology, using a combination of the texture and the image color information of the image, namely, the fractal features based on fractal dimension are selected to present the information texture, and the color features presented in RGB color space. These features are then used to train the network Kohonen, which will be represented by the underlying probability density function, the segmentation of this map is made by morphological watershed transformation. The performance of our color-texture segmentation approach is compared first, to color-based methods or texture-based methods only, and then to k-means method.

Keywords: Segmentation, color-texture, neural networks, fractal, watershed.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1377

Authors: Apoorva Vinod, Archana Mathur, Snehanshu Saha

Abstract:

Though there exists a plethora of Activation Functions (AFs) used in single and multiple hidden layer Neural Networks (NN), their behavior always raised curiosity, whether used in combination or singly. The popular AFs – Sigmoid, ReLU, and Tanh – have performed prominently well for shallow and deep architectures. Most of the time, AFs are used singly in multi-layered NN, and, to the best of our knowledge, their performance is never studied and analyzed deeply when used in combination. In this manuscript, we experiment on multi-layered NN architecture (both on shallow and deep architectures; Convolutional NN and VGG16) and investigate how well the network responds to using two different AFs (Sigmoid-Tanh, Tanh-ReLU, ReLU-Sigmoid) used alternately against a traditional, single (Sigmoid-Sigmoid, Tanh-Tanh, ReLU-ReLU) combination. Our results show that on using two different AFs, the network achieves better accuracy, substantially lower loss, and faster convergence on 4 computer vision (CV) and 15 Non-CV (NCV) datasets. When using different AFs, not only was the accuracy greater by 6-7%, but we also accomplished convergence twice as fast. We present a case study to investigate the probability of networks suffering vanishing and exploding gradients when using two different AFs. Additionally, we theoretically showed that a composition of two or more AFs satisfies Universal Approximation Theorem (UAT).

Keywords: Activation Function, Universal Approximation function, Neural Networks, convergence.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 161

Authors: Jianhua Wang, Yan Yu

Abstract:

This paper presents an intrusion detection system of hybrid neural network model based on RBF and Elman. It is used for anomaly detection and misuse detection. This model has the memory function .It can detect discrete and related aggressive behavior effectively. RBF network is a real-time pattern classifier, and Elman network achieves the memory ability for former event. Based on the hybrid model intrusion detection system uses DARPA data set to do test evaluation. It uses ROC curve to display the test result intuitively. After the experiment it proves this hybrid model intrusion detection system can effectively improve the detection rate, and reduce the rate of false alarm and fail.

Keywords: RBF, Elman, anomaly detection, misuse detection, hybrid neural network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2331

Authors: L. Salhi, M. Talbi, A. Cherif

Abstract:

This paper presents a new strategy of identification and classification of pathological voices using the hybrid method based on wavelet transform and neural networks. After speech acquisition from a patient, the speech signal is analysed in order to extract the acoustic parameters such as the pitch, the formants, Jitter, and shimmer. Obtained results will be compared to those normal and standard values thanks to a programmable database. Sounds are collected from normal people and patients, and then classified into two different categories. Speech data base is consists of several pathological and normal voices collected from the national hospital “Rabta-Tunis". Speech processing algorithm is conducted in a supervised mode for discrimination of normal and pathology voices and then for classification between neural and vocal pathologies (Parkinson, Alzheimer, laryngeal, dyslexia...). Several simulation results will be presented in function of the disease and will be compared with the clinical diagnosis in order to have an objective evaluation of the developed tool.

Keywords: Formants, Neural Networks, Pathological Voices, Pitch, Wavelet Transform.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2845

Authors: Jameela Ali Alkrimi, Hamid A. Jalab, Loay E. George, Abdul Rahim Ahmad, Azizah Suliman, Karim Al-Jashamy

Abstract:

Red blood cells (RBC) are the most common types of blood cells and are the most intensively studied in cell biology. The lack of RBCs is a condition in which the amount of hemoglobin level is lower than normal and is referred to as “anemia”. Abnormalities in RBCs will affect the exchange of oxygen. This paper presents a comparative study for various techniques for classifying the RBCs as normal or abnormal (anemic) using WEKA. WEKA is an open source consists of different machine learning algorithms for data mining applications. The algorithms tested are Radial Basis Function neural network, Support vector machine, and K-Nearest Neighbors algorithm. Two sets of combined features were utilized for classification of blood cells images. The first set, exclusively consist of geometrical features, was used to identify whether the tested blood cell has a spherical shape or non-spherical cells. While the second set, consist mainly of textural features was used to recognize the types of the spherical cells. We have provided an evaluation based on applying these classification methods to our RBCs image dataset which were obtained from Serdang Hospital - Malaysia, and measuring the accuracy of test results. The best achieved classification rates are 97%, 98%, and 79% for Support vector machines, Radial Basis Function neural network, and K-Nearest Neighbors algorithm respectively.

Keywords: K-Nearest Neighbors, Neural Network, Radial Basis Function, Red blood cells, Support vector machine.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3000

Authors: A. Muthuramalingam, S. Himavathi, E. Srinivasan

Abstract:

.Hardware realization of a Neural Network (NN), to a large extent depends on the efficient implementation of a single neuron. FPGA-based reconfigurable computing architectures are suitable for hardware implementation of neural networks. FPGA realization of ANNs with a large number of neurons is still a challenging task. This paper discusses the issues involved in implementation of a multi-input neuron with linear/nonlinear excitation functions using FPGA. Implementation method with resource/speed tradeoff is proposed to handle signed decimal numbers. The VHDL coding developed is tested using Xilinx XC V50hq240 Chip. To improve the speed of operation a lookup table method is used. The problems involved in using a lookup table (LUT) for a nonlinear function is discussed. The percentage saving in resource and the improvement in speed with an LUT for a neuron is reported. An attempt is also made to derive a generalized formula for a multi-input neuron that facilitates to estimate approximately the total resource requirement and speed achievable for a given multilayer neural network. This facilitates the designer to choose the FPGA capacity for a given application. Using the proposed method of implementation a neural network based application, namely, a Space vector modulator for a vector-controlled drive is presented

Keywords: FPGA implementation, multi-input neuron, neural network, nn based space vector modulator.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4434

Authors: Aziah Khamis, H. Shareef

Abstract:

The purpose of planned islanding is to construct a power island during system disturbances which are commonly formed for maintenance purpose. However, in most of the cases island mode operation is not allowed. Therefore distributed generators (DGs) must sense the unplanned disconnection from the main grid. Passive technique is the most commonly used method for this purpose. However, it needs improvement in order to identify the islanding condition. In this paper an effective method for identification of islanding condition based on phase space and neural network techniques has been developed. The captured voltage waveforms at the coupling points of DGs are processed to extract the required features. For this purposed a method known as the phase space techniques is used. Based on extracted features, two neural network configuration namely radial basis function and probabilistic neural networks are trained to recognize the waveform class. According to the test result, the investigated technique can provide satisfactory identification of the islanding condition in the distribution system.

Keywords: Classification, Islanding detection, Neural network, Phase space.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2135