scholarly journals Application of the neural network computing technology for calculating the interval-index characteristics of a minimally redundant modular code

2019 ◽  
Vol 62 (6) ◽  
pp. 652-660
Author(s):  
A. F. Chernyavsky ◽  
A. A. Kolyada ◽  
S. Yu. Protasenya
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
High Speed ◽  
Theoretical Estimate ◽  
Structural Complexity ◽  
Fold Increase ◽  
Number System ◽  
Parallel Structure ◽  
Network Computing ◽  
Time Expenditures

The article is devoted to the problem of creation of high-speed neural networks (NN) for calculation of interval-index characteristics of a minimally redundant modular code. The functional base of the proposed solution is an advanced class of neural networks of a final ring. These neural networks perform position-modular code transformations of scalable numbers using a modified reduction technology. A developed neural network has a uniform parallel structure, easy to implement and requires the time expenditures of the order (3[log2b]+ [log2k]+6tsum  close to the lower theoretical estimate. Here b and k is the average bit capacity and the number of modules respectively; t sum is the duration of the two-place operation of adding integers. The refusal from a normalization of the numbers of the modular code leads to a reduction of the required set of NN of the finite ring on the (k – 1) component. At the same time, the abnormal configuration of minimally redundant modular coding requires an average k-fold increase in the interval index module (relative to the rest of the bases of the modular number system). It leads to an adequate increase in hardware expenses on this module. Besides, the transition from normalized to unregulated coding reduces the level of homogeneity of the structure of the NN for calculating intervalindex characteristics. The possibility of reducing the structural complexity of the proposed NN by using abnormal intervalindex characteristics is investigated.

scholarly journals Implementation of Pruned Backpropagation Neural Network Based on Photonic Integrated Circuits

Photonics ◽  
2021 ◽  
Vol 8 (9) ◽  
pp. 363
Author(s):  
Qi Zhang ◽  
Zhuangzhuang Xing ◽  
Duan Huang
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Integrated Circuits ◽  
Energy Efficient ◽  
High Speed ◽  
Large Scale ◽  
Matrix Operation ◽  
Optical Neural Networks ◽  
Optical Neural Network ◽  
Random Initialization

We demonstrate a pruned high-speed and energy-efficient optical backpropagation (BP) neural network. The micro-ring resonator (MRR) banks, as the core of the weight matrix operation, are used for large-scale weighted summation. We find that tuning a pruned MRR weight banks model gives an equivalent performance in training with the model of random initialization. Results show that the overall accuracy of the optical neural network on the MNIST dataset is 93.49% after pruning six-layer MRR weight banks on the condition of low insertion loss. This work is scalable to much more complex networks, such as convolutional neural networks and recurrent neural networks, and provides a potential guide for truly large-scale optical neural networks.

A Review: Artificial Neural Network

2018 ◽  
Vol 8 (04) ◽  
Author(s):  
Jaspreet Kaur ◽  
Prabhpreet Kaur
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Network Architecture ◽  
High Speed ◽  
Parallel Architecture ◽  
Research Area ◽  
Neural Nets ◽  
Optimization Function ◽  
Artificial Neural

Neural networks are those information processing systems, which are built and performed to design the human brain. The main objective of the neural network research is to evolve a computational device for representing the brain to perform various evaluating tasks at a faster rate than the traditional systems. Neural networks are latest method of programming computers. Several programs that utilize neural nets are also proficient Neural networks have appeared in the past few years as an area of different opportunity for research area, development and application to a variety of real world problems because of their rapid feedback and parallel architecture. Artificial neural networks perform various tasks such as pattern-matching and classification, optimization function and data clustering. These tasks are very difficult for traditional for implementation of artificial neural networks, high-speed digital computers are used, which makes the simulation if neural processes feasible. This paper provides a broad overview of the wide array of artificial neural networks, some of the most commonly network architecture and various learning processes currently in use in research. Also concisely describes several applications of it.

Prediction of Surface Roughness in High Speed Milling Process Using the Artificial Neural Networks

2007 ◽  
Vol 364-366 ◽  
pp. 713-718 ◽  
Author(s):  
Dong Woo Kim ◽  
Young Jae Shin ◽  
Kyoung Taik Park ◽  
Eung Sug Lee ◽  
Jong Hyun Lee ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Surface Roughness ◽  
Artificial Neural Network ◽  
Artificial Neural Networks ◽  
High Speed ◽  
Back Propagation Algorithm ◽  
High Speed Milling ◽  
Artificial Neural ◽  
Cutting Path

The objective of this research was to apply the artificial neural network algorithm to predict the surface roughness in high speed milling operation. Tool length, feed rate, spindle speed, cutting path interval and run-out were used as five input neurons; and artificial neural networks model based on back-propagation algorithm was developed to predict the output neuron-surface roughness. A series of experiments was performed, and the results were estimated. The experimental results showed that the applied artificial neural network surface roughness prediction gave good accuracy in predicting the surface roughness under a variety of combinations of cutting conditions.

scholarly journals Digital Image Watermarking Processor Based on Deep Learning

Electronics ◽  
2021 ◽  
Vol 10 (10) ◽  
pp. 1183
Author(s):  
Jae-Eun Lee ◽  
Ji-Won Kang ◽  
Woo-Suk Kim ◽  
Jin-Kyum Kim ◽  
Young-Ho Seo ◽  
...  
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Integrated Circuit ◽  
High Speed ◽  
Learning Algorithm ◽  
Image Watermarking ◽  
Number System ◽  
Learning Technology ◽  
Optimization Methodology ◽  
Dedicated Processor

Much research and development have been made to implement deep neural networks for various purposes with hardware. We implement the deep learning algorithm with a dedicated processor. Watermarking technology for ultra-high resolution digital images and videos needs to be implemented in hardware for real-time or high-speed operation. We propose an optimization methodology to implement a deep learning-based watermarking algorithm in hardware. The proposed optimization methodology includes algorithm and memory optimization. Next, we analyze a fixed-point number system suitable for implementing neural networks as hardware for watermarking. Using these, a hardware structure of a dedicated processor for watermarking based on deep learning technology is proposed and implemented as an application-specific integrated circuit (ASIC).

scholarly journals Driver’s Seat Belt Detection Using CNN

2021 ◽  
Vol 12 (5) ◽  
pp. 776-785
Author(s):  
DS Bhupal Naik, G Sai Lakshmi, V Ramakrishna Sajja, D Venkatesulu,J Nageswara Rao
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Error Rate ◽  
High Speed ◽  
Deep Neural Network ◽  
Seat Belt ◽  
Transportation System ◽  
Convolution Neural Network ◽  
Classification Algorithms ◽  
Convolution Neural Networks

Seat belt detection is one of the necessary task which are required in transportation system to reduce accidents due to abrupt stop or high speed accident with other vehicles. In this paper, a technique is proposed to detect whether the driver wears seat belt or not by using convolution neural networks. Convolution Neural Network is nothing but deep Neural Network. ConvNet automatically collects features using filters or kernels from images without human involvement to classify the output images. Compared to different classification algorithms, preprocessing required in ConvNet is least. In this proposed method, first ConvNet is built and trained using Seatbelt dataset of both standard and non-standard. ConvNet learns the features from the images of seat belt dataset and performed better with an accuracy of 91.4% over SVM with 87.17% and an error rate of 8.55% when compared with SVM with 12.83% in case of standard dataset.

scholarly journals Optical coherent dot-product chip for sophisticated deep learning regression

2021 ◽  
Vol 10 (1) ◽  
Author(s):  
Shaofu Xu ◽  
Jing Wang ◽  
Haowen Shu ◽  
Zhike Zhang ◽  
Sicheng Yi ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Learning ◽  
Language Processing ◽  
High Speed ◽  
Autonomous Driving ◽  
Single Chip ◽  
Optical Fields ◽  
Large Bandwidth ◽  
Dot Product

AbstractOptical implementations of neural networks (ONNs) herald the next-generation high-speed and energy-efficient deep learning computing by harnessing the technical advantages of large bandwidth and high parallelism of optics. However, due to the problems of the incomplete numerical domain, limited hardware scale, or inadequate numerical accuracy, the majority of existing ONNs were studied for basic classification tasks. Given that regression is a fundamental form of deep learning and accounts for a large part of current artificial intelligence applications, it is necessary to master deep learning regression for further development and deployment of ONNs. Here, we demonstrate a silicon-based optical coherent dot-product chip (OCDC) capable of completing deep learning regression tasks. The OCDC adopts optical fields to carry out operations in the complete real-value domain instead of in only the positive domain. Via reusing, a single chip conducts matrix multiplications and convolutions in neural networks of any complexity. Also, hardware deviations are compensated via in-situ backpropagation control provided the simplicity of chip architecture. Therefore, the OCDC meets the requirements for sophisticated regression tasks and we successfully demonstrate a representative neural network, the AUTOMAP (a cutting-edge neural network model for image reconstruction). The quality of reconstructed images by the OCDC and a 32-bit digital computer is comparable. To the best of our knowledge, there is no precedent of performing such state-of-the-art regression tasks on ONN chips. It is anticipated that the OCDC can promote the novel accomplishment of ONNs in modern AI applications including autonomous driving, natural language processing, and scientific study.

scholarly journals Application of High-Speed Wind Tunnel Control Based on PID Neural Network

2011 ◽  
Vol 2-3 ◽  
pp. 3-6
Author(s):  
Ji Li ◽  
Hong Wang ◽  
Hai Long Huang
Keyword(s):  
Neural Network ◽  
Control System ◽  
Wind Tunnel ◽  
Pid Control ◽  
High Speed ◽  
Network Control ◽  
Parallel Structure ◽  
System Response ◽  
Back Propagation Algorithm ◽  
Pid Neural Network

The traditional PID control in nonlinear system such as high-speed wind tunnel has limitations, and the range of using is limited. The BP neural network has been widely applied to the optimization of the PID controller parameter adjustment. The PID neural network control system is introduced in the conventional PID control, which has advantages such as simple structure, physical meaning clear parameters, but also has a neural network of parallel structure and the function of learning and memory and nonlinear mapping capability. The controller uses BP (error back propagation) algorithm to correct connection weights, through on-line training and learning and make objective function to achieve optimal value. This improvement scheme can not only improve algorithm in the training process, and the convergence speed in the wind tunnel, the control valve control system response speed, high precision, meet the steady-state real-time control requirements.

scholarly journals Steel Surface Defect Classification Using Deep Residual Neural Network

Metals ◽  
2020 ◽  
Vol 10 (6) ◽  
pp. 846
Author(s):  
Ihor Konovalenko ◽  
Pavlo Maruschak ◽  
Janette Brezinová ◽  
Ján Viňáš ◽  
Jakub Brezina
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
High Speed ◽  
Steel Surface ◽  
Surface Defect ◽  
Surface Defects ◽  
Defect Classification ◽  
Flat Surfaces ◽  
Automated Method ◽  
Speed And Accuracy

An automated method for detecting and classifying three classes of surface defects in rolled metal has been developed, which allows for conducting defectoscopy with specified parameters of efficiency and speed. The possibility of using the residual neural networks for classifying defects has been investigated. The classifier based on the ResNet50 neural network is accepted as a basis. The model allows classifying images of flat surfaces with damage of three classes with the general accuracy of 96.91% based on the test data. The use of ResNet50 is shown to provide excellent recognition, high speed, and accuracy, which makes it an effective tool for detecting defects on metal surfaces.

scholarly journals Fast Approximations of Activation Functions in Deep Neural Networks when using Posit Arithmetic

Sensors ◽  
2020 ◽  
Vol 20 (5) ◽  
pp. 1515 ◽  
Author(s):  
Marco Cococcioni ◽  
Federico Rossi ◽  
Emanuele Ruffaldi ◽  
Sergio Saponara
Keyword(s):  
Neural Networks ◽  
Real Time ◽  
High Speed ◽  
Deep Neural Networks ◽  
Number System ◽  
Activation Function ◽  
Floating Point ◽  
Arithmetic Properties ◽  
Fast Processing ◽  
Point Representation

With increasing real-time constraints being put on the use of Deep Neural Networks (DNNs) by real-time scenarios, there is the need to review information representation. A very challenging path is to employ an encoding that allows a fast processing and hardware-friendly representation of information. Among the proposed alternatives to the IEEE 754 standard regarding floating point representation of real numbers, the recently introduced Posit format has been theoretically proven to be really promising in satisfying the mentioned requirements. However, with the absence of proper hardware support for this novel type, this evaluation can be conducted only through a software emulation. While waiting for the widespread availability of the Posit Processing Units (the equivalent of the Floating Point Unit (FPU)), we can already exploit the Posit representation and the currently available Arithmetic-Logic Unit (ALU) to speed up DNNs by manipulating the low-level bit string representations of Posits. As a first step, in this paper, we present new arithmetic properties of the Posit number system with a focus on the configuration with 0 exponent bits. In particular, we propose a new class of Posit operators called L1 operators, which consists of fast and approximated versions of existing arithmetic operations or functions (e.g., hyperbolic tangent (TANH) and extended linear unit (ELU)) only using integer arithmetic. These operators introduce very interesting properties and results: (i) faster evaluation than the exact counterpart with a negligible accuracy degradation; (ii) an efficient ALU emulation of a number of Posits operations; and (iii) the possibility to vectorize operations in Posits, using existing ALU vectorized operations (such as the scalable vector extension of ARM CPUs or advanced vector extensions on Intel CPUs). As a second step, we test the proposed activation function on Posit-based DNNs, showing how 16-bit down to 10-bit Posits represent an exact replacement for 32-bit floats while 8-bit Posits could be an interesting alternative to 32-bit floats since their performances are a bit lower but their high speed and low storage properties are very appealing (leading to a lower bandwidth demand and more cache-friendly code). Finally, we point out how small Posits (i.e., up to 14 bits long) are very interesting while PPUs become widespread, since Posit operations can be tabulated in a very efficient way (see details in the text).

Application of Convolutional Neural Networks for Automatic Number Plate Recognition on Complex Background Images

2015 ◽  
Vol 756 ◽  
pp. 695-703 ◽  
Author(s):  
A.A. Druki ◽  
J.A. Bolotova ◽  
V.G. Spitsyn
Keyword(s):  
Neural Network ◽  
Mathematical Modeling ◽  
Neural Networks ◽  
Convolutional Neural Network ◽  
Convolutional Neural Networks ◽  
High Speed ◽  
Character Segmentation ◽  
Pixel Intensity ◽  
Intensity Histogram ◽  
Complex Background

The relevance of this study is stipulated by the necessity of designing techniques, algorithms, and programs improving the efficiency of automatic number plate recognition (ANPR) on images with complex backgrounds.Purpose: The aim of this work is to improve the efficiency of automatic number plate recognition on images with complex backgrounds using methods, algorithms, and programs invariant to affine and projective transformations.Design/methodology: Such techniques as artificial intelligence, pattern identification and recognition, the theory of artificial neural networks (ANN), convolutional neural networks (CNN), evolutionary algorithms, mathematical modeling, the probability theory and mathematical statistics were applied via Visual Studio and MatLab software.Findings: The software is developed allowing the automatic number plate recognition on complex background images. The convolutional neural network comprising seven layers is suggested to identify the plate localization, i.e. finding and isolating the plate on the picture. The pixel intensity histogram-based algorithm was used for character segmentation or finding individual characters on the plates. The convolutional neural network comprising six layers is designed to recognize characters. The suggested software system allows automatic number plate recognition at large angles of inclinations and rather a high speed.

Sign in / Sign up

Export Citation Format

Share Document