Search results for: neural cellular automata

Authors: Ghassan A. Abed, Samir I. Badrawi

Abstract:

The advancement in wireless technology with the wide use of mobile devices have drawn the attention of the research and technological communities towards wireless environments, such as Wireless Local Area Networks (WLANs), Wireless Wide Area Networks (WWANs), and mobile systems and ad-hoc networks. Unfortunately, wired and wireless networks are expressively different in terms of link reliability, bandwidth, and time of propagation delay and by adapting new solutions for these enhanced telecommunications, superior quality, efficiency, and opportunities will be provided where wireless communications were otherwise unfeasible. Some researchers define 4G as a significant improvement of 3G, where current cellular network’s issues will be solved and data transfer will play a more significant role. For others, 4G unifies cellular and wireless local area networks, and introduces new routing techniques, efficient solutions for sharing dedicated frequency bands, and an increased mobility and bandwidth capacity. This paper discusses the possible solutions and enhancements probabilities that proposed to improve the performance of Transmission Control Protocol (TCP) over different wireless networks and also the paper investigated each approach in term of advantages and disadvantages.

Keywords: TCP, Wireless Networks, Cellular Systems, WLAN.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1927

Authors: Prof. Chokri SLIM

Abstract:

A Novel fuzzy neural network combining with support vector learning mechanism called support-vector-based fuzzy neural networks (SVBFNN) is proposed. The SVBFNN combine the capability of minimizing the empirical risk (training error) and expected risk (testing error) of support vector learning in high dimensional data spaces and the efficient human-like reasoning of FNN.

Keywords: Neural network, fuzzy inference, machine learning, fuzzy modeling and rule extraction, support vector regression.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 16682

Authors: Jasmin Velagic, Nedim Osmic, Bakir Lacevic

Abstract:

In this paper the neural network-based controller is designed for motion control of a mobile robot. This paper treats the problems of trajectory following and posture stabilization of the mobile robot with nonholonomic constraints. For this purpose the recurrent neural network with one hidden layer is used. It learns relationship between linear velocities and error positions of the mobile robot. This neural network is trained on-line using the backpropagation optimization algorithm with an adaptive learning rate. The optimization algorithm is performed at each sample time to compute the optimal control inputs. The performance of the proposed system is investigated using a kinematic model of the mobile robot.

Keywords: Mobile robot, kinematic model, neural network, motion control, adaptive learning rate.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3322

Authors: Paul Lajbcygier, Seng Lee

Abstract:

Co-integration models the long-term, equilibrium relationship of two or more related financial variables. Even if cointegration is found, in the short run, there may be deviations from the long run equilibrium relationship. The aim of this work is to forecast these deviations using neural networks and create a trading strategy based on them. A case study is used: co-integration residuals from Australian Bank Bill futures are forecast and traded using various exogenous input variables combined with neural networks. The choice of the optimal exogenous input variables chosen for each neural network, undertaken in previous work [1], is validated by comparing the forecasts and corresponding profitability of each, using a trading strategy.

Keywords: Artificial neural networks, co-integration, forecasting, trading rule.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1237

Authors: P. R. Jayathilaka, K.L. Jayaratne, H.L. Premaratne

Abstract:

This research presented in this paper is an on-going project of an application of neural network and fuzzy models to evaluate the sociological factors which affect the educational performance of the students in Sri Lanka. One of its major goals is to prepare the grounds to device a counseling tool which helps these students for a better performance at their examinations, especially at their G.C.E O/L (General Certificate of Education-Ordinary Level) examination. Closely related sociological factors are collected as raw data and the noise of these data are filtered through the fuzzy interface and the supervised neural network is being utilized to recognize the performance patterns against the chosen social factors.

Keywords: Education, Fuzzy, neural network, prediction, Sociology

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1629

Authors: N.Vahdatifar, Ss.Mortazavi, R.Kianinezhad

Abstract:

In this paper, a Neural Network based predictive DTC algorithm is proposed .This approach is used as an alternative to classical approaches .An appropriate riate Feed - forward network is chosen and based on its value of derivative electromagnetic torque ; optimal stator voltage vector is determined to be applied to the induction motor (by inverter). Moreover, an appropriate torque and flux observer is proposed.

Keywords: Neural Networks, Predictive DTC

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1376

Authors: Li Kewen, Su Zhaoxin, Wang Xingmou, Zhu Jian Bing

Abstract:

Conventional reservoir prediction methods ar not sufficient to explore the implicit relation between seismic attributes, and thus data utilization is low. In order to improve the predictive classification accuracy of reservoir lithology, this paper proposes a deep learning lithology prediction method based on ResNet (Residual Neural Network) and SENet (Squeeze-and-Excitation Neural Network). The neural network model is built and trained by using seismic attribute data and lithology data of Shengli oilfield, and the nonlinear mapping relationship between seismic attribute and lithology marker is established. The experimental results show that this method can significantly improve the classification effect of reservoir lithology, and the classification accuracy is close to 70%. This study can effectively predict the lithology of undrilled area and provide support for exploration and development.

Keywords: Convolutional neural network, lithology, prediction of reservoir lithology, seismic attributes.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 643

Authors: Liyuan Wang, Shouming Zhong

Abstract:

In this paper, the design problem of state estimator for neural networks with the mixed time-varying delays are investigated by constructing appropriate Lyapunov-Krasovskii functionals and using some effective mathematical techniques. In order to derive several conditions to guarantee the estimation error systems to be globally exponential stable, we transform the considered systems into the neural-type time-delay systems. Then with a set of linear inequalities(LMIs), we can obtain the stable criteria. Finally, three numerical examples are given to show the effectiveness and less conservatism of the proposed criterion.

Keywords: State estimator, Neural networks, Globally exponential stability.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1656

Authors: Magdi M. Nabi, Ding-Li Yu

Abstract:

Control of a semi-batch polymerization reactor using an adaptive radial basis function (RBF) neural network method is investigated in this paper. A neural network inverse model is used to estimate the valve position of the reactor; this method can identify the controlled system with the RBF neural network identifier. The weights of the adaptive PID controller are timely adjusted based on the identification of the plant and self-learning capability of RBFNN. A PID controller is used in the feedback control to regulate the actual temperature by compensating the neural network inverse model output. Simulation results show that the proposed control has strong adaptability, robustness and satisfactory control performance and the nonlinear system is achieved.

Keywords: Chylla-Haase polymerization reactor, RBF neural networks, feed-forward and feedback control.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2667

Authors: Ning Xue, Srinivas Merugu, Ignacio Delgado Martinez, Tao Sun, John Tsang, Shih-Cheng Yen

Abstract:

We have developed a polyimide based neural interface electrode to record nerve signals from the sciatic nerve of a rat. The neural interface electrode has a split-ring shape, with four protruding gold electrodes for recording, and two reference gold electrodes around the split-ring. The split-ring electrode can be opened up to encircle the sciatic nerve. The four electrodes can be bent to sit on top of the nerve and hold the device in position, while the split-ring frame remains flat. In comparison, while traditional cuff electrodes can only fit certain sizes of the nerve, the developed device can fit a variety of rat sciatic nerve dimensions from 0.6 mm to 1.0 mm, and adapt to the chronic changes in the nerve as the electrode tips are bendable. The electrochemical impedance spectroscopy measurement was conducted. The gold electrode impedance is on the order of 10 kΩ, showing excellent charge injection capacity to record neural signals.

Keywords: Impedance, neural interface, split-ring electrode.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2588

Authors: Mohsen Hayati, Zahra Mohebi

Abstract:

In this paper, the application of neural networks to study the design of short-term temperature forecasting (STTF) Systems for Kermanshah city, west of Iran was explored. One important architecture of neural networks named Multi-Layer Perceptron (MLP) to model STTF systems is used. Our study based on MLP was trained and tested using ten years (1996-2006) meteorological data. The results show that MLP network has the minimum forecasting error and can be considered as a good method to model the STTF systems.

Keywords: Artificial neural networks, Forecasting, Weather, Multi-layer perceptron.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 4344

Authors: Meng Hu, Lili Wang

Abstract:

By using the method of coincidence degree and constructing suitable Lyapunov functional, some sufficient conditions are established for the existence and global exponential stability of antiperiodic solutions for a kind of impulsive Cohen-Grossberg shunting inhibitory cellular neural networks (CGSICNNs) on time scales. An example is given to illustrate our results.

Keywords: Anti-periodic solution, coincidence degree, CGSICNNs, impulse, time scales.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1300

Authors: Hazem M. El-Bakry

Abstract:

Neural processors have shown good results for detecting a certain character in a given input matrix. In this paper, a new idead to speed up the operation of neural processors for character detection is presented. Such processors are designed based on cross correlation in the frequency domain between the input matrix and the weights of neural networks. This approach is developed to reduce the computation steps required by these faster neural networks for the searching process. The principle of divide and conquer strategy is applied through image decomposition. Each image is divided into small in size sub-images and then each one is tested separately by using a single faster neural processor. Furthermore, faster character detection is obtained by using parallel processing techniques to test the resulting sub-images at the same time using the same number of faster neural networks. In contrast to using only faster neural processors, the speed up ratio is increased with the size of the input image when using faster neural processors and image decomposition. Moreover, the problem of local subimage normalization in the frequency domain is solved. The effect of image normalization on the speed up ratio of character detection is discussed. Simulation results show that local subimage normalization through weight normalization is faster than subimage normalization in the spatial domain. The overall speed up ratio of the detection process is increased as the normalization of weights is done off line.

Keywords: Fast Character Detection, Neural Processors, Cross Correlation, Image Normalization, Parallel Processing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1529

Authors: Riad S. Jabri

Abstract:

Pattern matching based on regular tree grammars have been widely used in many areas of computer science. In this paper, we propose a pattern matcher within the framework of code generation, based on a generic and a formalized approach. According to this approach, parsers for regular tree grammars are adapted to a general pattern matching solution, rather than adapting the pattern matching according to their parsing behavior. Hence, we first formalize the construction of the pattern matches respective to input trees drawn from a regular tree grammar in a form of the so-called match trees. Then, we adopt a recently developed generic parser and tightly couple its parsing behavior with such construction. In addition to its generality, the resulting pattern matcher is characterized by its soundness and efficient implementation. This is demonstrated by the proposed theory and by the derived algorithms for its implementation. A comparison with similar and well-known approaches, such as the ones based on tree automata and LR parsers, has shown that our pattern matcher can be applied to a broader class of grammars, and achieves better approximation of pattern matches in one pass. Furthermore, its use as a machine code selector is characterized by a minimized overhead, due to the balanced distribution of the cost computations into static ones, during parser generation time, and into dynamic ones, during parsing time.

Keywords: Bottom-up automata, Code selection, Pattern matching, Regular tree grammars, Match trees.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1261

Authors: P. Karimi, A. H. Khedmati Bazkiaei

Abstract:

The main goal of this study is to test differential neural network as a controller of smart structure and is to enumerate its advantages and disadvantages in comparison with other controllers. In this study, the smart structure has been considered as a Euler Bernoulli cantilever beam and it has been tried that it be under control with the use of vibration neural network resulting from movement. Also, a linear observer has been considered as a reference controller and has been compared its results. The considered vibration charts and the controlled state have been recounted in the final part of this text. The obtained result show that neural observer has better performance in comparison to the implemented linear observer.

Keywords: Smart material, on-line differential artificial neural network, active control, finite element method.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 807

Authors: Chao Wang, Yongkun Li

Abstract:

In this paper, the discrete-time fuzzy BAM neural network with delays and impulses is studied. Sufficient conditions are obtained for the existence and global stability of a unique equilibrium of this class of fuzzy BAM neural networks with Lipschitzian activation functions without assuming their boundedness, monotonicity or differentiability and subjected to impulsive state displacements at fixed instants of time. Some numerical examples are given to demonstrate the effectiveness of the obtained results.

Keywords: Discrete-time fuzzy BAM neural networks, ımpulses, global exponential stability, global asymptotical stability, equilibrium point.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1495

Authors: Mohammad Saleh Vahdatpour, Yanqing Zhang

Abstract:

Understanding the neural mechanisms underlying motion detection in the human visual system has long been a fascinating challenge in neuroscience and artificial intelligence. This paper presents a spiking neural network model inspired by the processing of motion information in the primate visual system, particularly focusing on the Middle Temporal (MT) area. In our study, we propose a multi-layer spiking neural network model to perform motion detection tasks, leveraging the idea that synaptic delays in neuronal communication are pivotal in motion perception. Synaptic delay, determined by factors like axon length and myelin insulation, affects the temporal order of input spikes, thereby encoding motion direction and speed. Overall, our spiking neural network model demonstrates the feasibility of capturing motion detection principles observed in the primate visual system. The combination of synaptic delays, learning mechanisms, and shared weights and delays in SMD provides a promising framework for motion perception in artificial systems, with potential applications in computer vision and robotics.

Keywords: Neural networks, motion detection, signature detection, convolutional neural network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 138

Authors: E. Nemati Lay, M. Peymani, E. Sanjari

Abstract:

Prediction of viscosity of natural gas is an important parameter in the energy industries such as natural gas storage and transportation. In this study viscosity of different compositions of natural gas is modeled by using an artificial neural network (ANN) based on back-propagation method. A reliable database including more than 3841 experimental data of viscosity for testing and training of ANN is used. The designed neural network can predict the natural gas viscosity using pseudo-reduced pressure and pseudo-reduced temperature with AARD% of 0.221. The accuracy of designed ANN has been compared to other published empirical models. The comparison indicates that the proposed method can provide accurate results.

Keywords: Artificial neural network, Empirical correlation, Natural gas, Viscosity

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3230

Authors: Hyun-Jee Yang, Bit-Na Kwon, Yong-Jun Kim, Hyoung-Kyu Song

Abstract:

In the cooperative transmission scheme, both the cellular system and broadcasting system are composed. Two cellular base stations (CBSs) communicating with a user in the cell edge use cooperative transmission scheme in the conventional scheme. In the case that the distance between two CBSs and the user is distant, the conventional scheme does not guarantee the quality of the communication because the channel condition is bad. Therefore, if the distance between CBSs and a user is distant, the performance of the conventional scheme is decreased. Also, the bad channel condition has bad effects on the performance. The proposed scheme uses two relays to communicate well with CBSs when the channel condition between CBSs and the user is poor. Using the relay in the high attenuation environment can obtain both advantages of the high bit error rate (BER) and throughput performance.

Keywords: Cooperative communications, diversity gain, OFDM, interworking system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1746

Authors: Si Mon Kueh, Tom J. Kazmierski

Abstract:

There are about 1% of the world population suffering from the hidden disability known as epilepsy and major developing countries are not fully equipped to counter this problem. In order to reduce the inconvenience and danger of epilepsy, different methods have been researched by using a artificial neural network (ANN) classification to distinguish epileptic waveforms from normal brain waveforms. This paper outlines the aim of achieving massive ANN parallelization through a dedicated hardware using bit-serial processing. The design of this bit-serial Neural Processing Element (NPE) is presented which implements the functionality of a complete neuron using variable accuracy. The proposed design has been tested taking into consideration non-idealities of a hardware ANN. The NPE consists of a bit-serial multiplier which uses only 16 logic elements on an Altera Cyclone IV FPGA and a bit-serial ALU as well as a look-up table. Arrays of NPEs can be driven by a single controller which executes the neural processing algorithm. In conclusion, the proposed compact NPE design allows the construction of complex hardware ANNs that can be implemented in a portable equipment that suits the needs of a single epileptic patient in his or her daily activities to predict the occurrences of impending tonic conic seizures.

Keywords: Artificial Neural Networks, bit-serial neural processor, FPGA, Neural Processing Element.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1564

Authors: Ersin Gose, Kenan Buyukatak, Onur Osman, Osman N. Ucan

Abstract:

To compress, improve bit error performance and also enhance 2D images, a new scheme, called Iterative Cellular-Turbo System (IC-TS) is introduced. In IC-TS, the original image is partitioned into 2N quantization levels, where N is denoted as bit planes. Then each of the N-bit-plane is coded by Turbo encoder and transmitted over Additive White Gaussian Noise (AWGN) channel. At the receiver side, bit-planes are re-assembled taking into consideration of neighborhood relationship of pixels in 2-D images. Each of the noisy bit-plane values of the image is evaluated iteratively using IC-TS structure, which is composed of equalization block; Iterative Cellular Image Processing Algorithm (ICIPA) and Turbo decoder. In IC-TS, there is an iterative feedback link between ICIPA and Turbo decoder. ICIPA uses mean and standard deviation of estimated values of each pixel neighborhood. It has extra-ordinary satisfactory results of both Bit Error Rate (BER) and image enhancement performance for less than -1 dB Signal-to-Noise Ratio (SNR) values, compared to traditional turbo coding scheme and 2-D filtering, applied separately. Also, compression can be achieved by using IC-TS systems. In compression, less memory storage is used and data rate is increased up to N-1 times by simply choosing any number of bit slices, sacrificing resolution. Hence, it is concluded that IC-TS system will be a compromising approach in 2-D image transmission, recovery of noisy signals and image compression.

Keywords: Iterative Cellular Image Processing Algorithm (ICIPA), Turbo Coding, Iterative Cellular Turbo System (IC-TS), Image Compression.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1804

Authors: John Kabuba, Antoine Mulaba-Bafubiandi, Kim Battle

Abstract:

In this study a neural network (NN) was proposed to predict the sorption of binary mixture of copper-cobalt ions into clinoptilolite as ion-exchanger. The configuration of the backpropagation neural network giving the smallest mean square error was three-layer NN with tangent sigmoid transfer function at hidden layer with 10 neurons, linear transfer function at output layer and Levenberg-Marquardt backpropagation training algorithm. Experiments have been carried out in the batch reactor to obtain equilibrium data of the individual sorption and the mixture of coppercobalt ions. The obtained modeling results have shown that the used of neural network has better adjusted the equilibrium data of the binary system when compared with the conventional sorption isotherm models.

Keywords: Adsorption isotherm, binary system, neural network; sorption

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2035

Authors: Jaswinder Kaur, Satwinder Singh, Karanjeet Singh Kahlon

Abstract:

Accurate software cost estimates are critical to both developers and customers. They can be used for generating request for proposals, contract negotiations, scheduling, monitoring and control. The exact relationship between the attributes of the effort estimation is difficult to establish. A neural network is good at discovering relationships and pattern in the data. So, in this paper a comparative analysis among existing Halstead Model, Walston-Felix Model, Bailey-Basili Model, Doty Model and Neural Network Based Model is performed. Neural Network has outperformed the other considered models. Hence, we proposed Neural Network system as a soft computing approach to model the effort estimation of the software systems.

Keywords: Effort Estimation, Neural Network, Halstead Model, Walston-Felix Model, Bailey-Basili Model, Doty Model.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2214

Authors: Liu Zhiyuan, Sun Zongdi

Abstract:

In this paper, the BP neural network model is established to predict the carbon trading price and carbon trading volume in Shanghai City. First of all, we find the data of carbon trading price and carbon trading volume in Shanghai City from September 30, 2015 to December 23, 2016. The carbon trading price and trading volume data were processed to get the average value of each 5, 10, 20, 30, and 60 carbon trading price and trading volume. Then, these data are used as input of BP neural network model. Finally, after the training of BP neural network, the prediction values of Shanghai carbon trading price and trading volume are obtained, and the model is tested.

Keywords: Carbon trading price, carbon trading volume, BP neural network model, Shanghai City.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1385

Authors: Raj Kumar Samanta, Partha Bhattacharjee Gautam Sanyal

Abstract:

In this paper, a mathematical model is proposed to estimate the dropping probabilities of cellular wireless networks by queuing handoff instead of reserving guard channels. Usually, prioritized handling of handoff calls is done with the help of guard channel reservation. To evaluate the proposed model, gamma inter-arrival and general service time distributions have been considered. Prevention of some of the attempted calls from reaching to the switching center due to electromagnetic propagation failure or whimsical user behaviour (missed call, prepaid balance etc.), make the inter-arrival time of the input traffic to follow gamma distribution. The performance is evaluated and compared with that of guard channel scheme.

Keywords: Cellular wireless networks, non-classical traffic, mathematicalmodel, guard channel, queuing, handoff.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2384

Authors: Baeza S. Roberto

Abstract:

The knitted fabric suffers a deformation in its dimensions due to stretching and tension factors, transverse and longitudinal respectively, during the process in rectilinear knitting machines so it performs a dry relaxation shrinkage procedure and thermal action of prefixed to obtain stable conditions in the knitting. This paper presents a dry relaxation shrinkage prediction of Bordeaux fiber using a feed forward neural network and linear regression models. Six operational alternatives of shrinkage were predicted. A comparison of the results was performed finding neural network models with higher levels of explanation of the variability and prediction. The presence of different reposes is included. The models were obtained through a neural toolbox of Matlab and Minitab software with real data in a knitting company of Southern Guanajuato. The results allow predicting dry relaxation shrinkage of each alternative operation.

Keywords: Neural network, dry relaxation, knitting, linear regression.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1751

Authors: Duy M. P. Vo, Cornelia Sennewald, Gerald Hoffmann, Chokri Cherif

Abstract:

The development of solutions to improve the resistance of buildings to short-term dynamic loads, particularly impact load, is driven by the urgent demand worldwide on securing human life and critical infrastructures. The research training group GRK 2250/1 aims to develop mineral-bonded composites that allow the fabrication of thin-layered strengthening layers providing available concrete members with enhanced impact resistance. This paper presents the development of 3D woven wire cellular structures that can be used as innovative reinforcement for targeted composites. 3D woven wire cellular structures are truss-like architectures that can be fabricated in an automatized process with a great customization possibility. The specific architecture allows this kind of structures to have good load bearing capability and forming behavior, which is of great potential to give strength against impact loading. An appropriate combination of topology and material enables an optimal use of thin-layered reinforcement in concrete constructions.

Keywords: 3D woven cellular structures, ductile behavior, energy absorption, fiber-based reinforced concrete, impact resistant.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 688

Authors: Alexander Goncharenko, Andrey Denisov, Sergey Alyamkin, Evgeny Terentev

Abstract:

The neural network quantization is highly desired procedure to perform before running neural networks on mobile devices. Quantization without fine-tuning leads to accuracy drop of the model, whereas commonly used training with quantization is done on the full set of the labeled data and therefore is both time- and resource-consuming. Real life applications require simplification and acceleration of quantization procedure that will maintain accuracy of full-precision neural network, especially for modern mobile neural network architectures like Mobilenet-v1, MobileNet-v2 and MNAS. Here we present a method to significantly optimize training with quantization procedure by introducing the trained scale factors for discretization thresholds that are separate for each filter. Using the proposed technique, we quantize the modern mobile architectures of neural networks with the set of train data of only ∼ 10% of the total ImageNet 2012 sample. Such reduction of train dataset size and small number of trainable parameters allow to fine-tune the network for several hours while maintaining the high accuracy of quantized model (accuracy drop was less than 0.5%). Ready-for-use models and code are available in the GitHub repository.

Keywords: Distillation, machine learning, neural networks, quantization.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 722

Authors: Kexin Jia, Youxin Lu

Abstract:

To make the modulation classification system more suitable for signals in a wide range of signal to noise rate (SNR), a novel method of designing combined classifier based on fuzzy neural network (FNN) is presented in this paper. The method employs fuzzy neural network classifiers and interclass distance (ICD) to improve recognition reliability. Experimental results show that the proposed combined classifier has high recognition rate with large variation range of SNR (success rates are over 99.9% when SNR is not lower than 5dB).

Keywords: Modulation classification, combined classifier, fuzzy neural network, interclass distance.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1213

Authors: Sorin Valcan, Mihail Găianu

Abstract:

Labeling is a very costly and time consuming process which aims to generate datasets for training neural networks in several functionalities and projects. For driver monitoring system projects, the need of labeled images has a significant impact on the budget and distribution of effort. This paper presents the modifications done to a ground truth data generation algorithm for 2D eyes location on infrared images with drivers in order to improve the quality of the data and performance of the trained neural networks. The algorithm restrictions become tougher which makes it more accurate but also less constant. The resulting dataset becomes smaller and shall not be altered by any kind of manual labels adjustment before being used in the neural networks training process. These changes resulted in a much better performance of the trained neural networks.

Keywords: Labeling automation, infrared camera, driver monitoring, eye detection, Convolutional Neural Networks.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 398