scholarly journals An On-Chip Learning Method for Neuromorphic Systems Based on Non-Ideal Synapse Devices

Electronics ◽  
2020 ◽  
Vol 9 (11) ◽  
pp. 1946
Author(s):  
Jae-Eun Lee ◽  
Chuljun Lee ◽  
Dong-Wook Kim ◽  
Daeseok Lee ◽  
Young-Ho Seo
Keyword(s):  
Neural Network ◽  
System Performance ◽  
Learning Method ◽  
Ideal Case ◽  
The Neural Network ◽  
Neuromorphic Systems ◽  
Conductance Changes ◽  
Neuromorphic System ◽  
On Chip ◽  
Conductance Level

In this paper, we propose an on-chip learning method that can overcome the poor characteristics of pre-developed practical synaptic devices, thereby increasing the accuracy of the neural network based on the neuromorphic system. The fabricated synaptic devices, based on Pr1−xCaxMnO3, LiCoO2, and TiOx, inherently suffer from undesirable characteristics, such as nonlinearity, discontinuities, and asymmetric conductance responses, which degrade the neuromorphic system performance. To address these limitations, we have proposed a conductance-based linear weighted quantization method, which controls conductance changes, and trained a neural network to predict the handwritten digits from the standard database MNIST. Furthermore, we quantitatively considered the non-ideal case, to ensure reliability by limiting the conductance level to that which synaptic devices can practically accept. Based on this proposed learning method, we significantly improved the neuromorphic system, without any hardware modifications to the synaptic devices or neuromorphic systems. Thus, the results emphatically show that, even for devices with poor synaptic characteristics, the neuromorphic system performance can be improved.

Pattern-Based Deep Learning Method to Extract Information from the Log Dataset

2021 ◽  
pp. 2150296
Author(s):  
Xi Li ◽  
Ting Wang ◽  
Shexiong Wang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Data Sets ◽  
Learning Method ◽  
Statistical Features ◽  
Log Data ◽  
The Neural Network ◽  
Testing Data ◽  
Extract Information

It draws researchers’ attentions how to make use of the log data effectively without paying much for storing them. In this paper, we propose pattern-based deep learning method to extract the features from log datasets and to facilitate its further use at the reasonable expense of the storage performances. By taking the advantages of the neural network and thoughts to combine statistical features with experts’ knowledge, there are satisfactory results in the experiments on some specified datasets and on the routine systems that our group maintains. Processed on testing data sets, the model is 5%, at least, more likely to outperform its competitors in accuracy perspective. More importantly, its schema unveils a new way to mingle experts’ experiences with statistical log parser.

Cluster Based Medical Image Registration Using Optimized Neural Network

2017 ◽  
pp. 1437-1467
Author(s):  
Joydev Hazra ◽  
Aditi Roy Chowdhury ◽  
Paramartha Dutta
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Image Registration ◽  
Convergence Rate ◽  
Optimization Algorithm ◽  
Clustering Algorithms ◽  
Data Sets ◽  
Learning Method ◽  
Local Minima ◽  
The Neural Network

Registration of medical images like CT-MR, MR-MR etc. are challenging area for researchers. This chapter introduces a new cluster based registration technique with help of the supervised optimized neural network. Features are extracted from different cluster of an image obtained from clustering algorithms. To overcome the drawback regarding convergence rate of neural network, an optimized neural network is proposed in this chapter. The weights are optimized to increase the convergence rate as well as to avoid stuck in local minima. Different clustering algorithms are explored to minimize the clustering error of an image and extract features from suitable one. The supervised learning method applied to train the neural network. During this training process an optimization algorithm named Genetic Algorithm (GA) is used to update the weights of a neural network. To demonstrate the effectiveness of the proposed method, investigation is carried out on MR T1, T2 data sets. The proposed method shows convincing results in comparison with other existing techniques.

scholarly journals Toward an embedded system for gesture recognition based on artificial neural network using reconfigurable target (case study and review)

2020 ◽  
Vol 10 (4) ◽  
pp. 142-148
Author(s):  
Ahmad Reda ◽  
Tareq Alshoufi ◽  
Ahmed Bouzid ◽  
József Vásárhelyi
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Embedded System ◽  
Time Series Data ◽  
Reference Signal ◽  
Series Data ◽  
The Neural Network ◽  
Human Arm ◽  
On Chip

With a view to create an intelligent remote control for robot movements, this article treats the study case of dataset creation using RSG (Reference Signal Generator). Using artificial intelligence, the device recognizes the gestures of an operator. Indeed, a neural network can classify time series data coming from accelerometers, and for a beginning 4 gestures are taken into consideration. The most challenging work is to build a reference dataset that is necessary for the learning process. To train the neural network, a huge amount of reference data should be created (hundreds of thousands of time-series vectors per gesture per sensor), which cannot be done manually by an operator. To overcome the issue, an RSG is created. This article also describes how a 1-DoF arm has been designed to emulate the behavior of the human arm doing gestures as well as the data acquisition system. The system is based on a software/hardware co-design implemented on Programmable System on Chip (PSoC).

scholarly journals Construction of a neural network energy function for protein physics

2021 ◽  
Author(s):  
Huan Yang ◽  
Zhaoping Xiong ◽  
Francesco Zonta
Keyword(s):  
Neural Network ◽  
Experimental Data ◽  
Neural Networks ◽  
Energy Function ◽  
Protein Sequence ◽  
Deep Neural Networks ◽  
Protein Structures ◽  
Learning Method ◽  
Local Structures ◽  
The Neural Network

AbstractClassical potentials are widely used to describe protein physics, due to their simplicity and accuracy, but they are continuously challenged as real applications become more demanding with time. Deep neural networks could help generating alternative ways of describing protein physics. Here we propose an unsupervised learning method to derive a neural network energy function for proteins. The energy function is a probability density model learned from plenty of 3D local structures which have been extensively explored by evolution. We tested this model on a few applications (assessment of protein structures, protein dynamics and protein sequence design), showing that the neural network can correctly recognize patterns in protein structures. In other words, the neural network learned some aspects of protein physics from experimental data.

scholarly journals EMERGENCY CONTROL SYSTEM BASED ON NEURAL NETWORKS AND FUZZY LOGIC

2020 ◽  
Vol 1 (1) ◽  
pp. 45-52
Author(s):  
S.M. Konovalov ◽  
◽  
G.A. Yegoshyna ◽  
S.M. Voronoy ◽  
Keyword(s):  
Neural Network ◽  
Fuzzy Logic ◽  
Expert System ◽  
Knowledge Base ◽  
Performance Indicators ◽  
System Performance ◽  
Operation Mode ◽  
Operational Reliability ◽  
Efficiency Coefficient ◽  
The Neural Network

The presented paper investigates the problem of ensuring the safety of modern vessels, represented as complex organizational and technical systems. This study solves the task of diagnosing and predicting the level of ships’ operational reliability using a hybrid expert system based on a combination of a neural network and fuzzy logic. Trends in modern control systems show that they must be adaptive and intelligent. However, these requirements cannot be met by expert systems based only on fuzzy logic. This work explores the possibility of combining neural network modules with fuzzy logic and considers the features of emergency management stages based on the offered hybrid expert system. The input information arrives in a knowledge base through gauges, where it is structured and distributed in the form of performance indicators. Emergency recommendations for the operator are formed as a result of a combination of performance indicators available in the knowledge base. Modules of the neural network and fuzzy logic form a system for assessing a complex technical system’s health based on calculated estimates of the health of technical nodes. In addition, the authors formed a hierarchy of factors affecting the reliability of the system. While developing the knowledge base, critical values for each variable influencing the system performance are set, and when the values are reached, the operation mode becomes an emergency. The authors chose a multilayer perceptron with a layer of recurrent neurons and inputs as fed factors and criteria for performance; one output displays the value of system performance. Prediction of the technical state of the system is made based on time series analysis. The system with six variables was used as a test set, three of which are non-linguistic (efficiency coefficient, temperature, and pressure). The standard linguistic variable, calculated by the neural network, includes speed, fuel consumption, and wear of the node. The fuzzy logic module was used to form recommendations for the prevention or elimination of an emergency.

Cluster Based Medical Image Registration Using Optimized Neural Network

2016 ◽  
pp. 551-581 ◽  
Author(s):  
Joydev Hazra ◽  
Aditi Roy Chowdhury ◽  
Paramartha Dutta
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Image Registration ◽  
Convergence Rate ◽  
Optimization Algorithm ◽  
Clustering Algorithms ◽  
Data Sets ◽  
Learning Method ◽  
Local Minima ◽  
The Neural Network

Registration of medical images like CT-MR, MR-MR etc. are challenging area for researchers. This chapter introduces a new cluster based registration technique with help of the supervised optimized neural network. Features are extracted from different cluster of an image obtained from clustering algorithms. To overcome the drawback regarding convergence rate of neural network, an optimized neural network is proposed in this chapter. The weights are optimized to increase the convergence rate as well as to avoid stuck in local minima. Different clustering algorithms are explored to minimize the clustering error of an image and extract features from suitable one. The supervised learning method applied to train the neural network. During this training process an optimization algorithm named Genetic Algorithm (GA) is used to update the weights of a neural network. To demonstrate the effectiveness of the proposed method, investigation is carried out on MR T1, T2 data sets. The proposed method shows convincing results in comparison with other existing techniques.

Development of Neural-Network Interatomic Potentials for Structural Materials

2016 ◽  
Vol 258 ◽  
pp. 69-72
Author(s):  
Ryo Kobayashi ◽  
Tomoyuki Tamura ◽  
Ichiro Takeuchi ◽  
Shuji Ogata
Keyword(s):  
Neural Network ◽  
Machine Learning ◽  
Md Simulation ◽  
Interatomic Potential ◽  
Basis Functions ◽  
Interatomic Potentials ◽  
Machine Learning Method ◽  
Learning Method ◽  
New Approach ◽  
The Neural Network

The validity of the molecular dynamics (MD) simulation is highly dependent on the accuracy or reproducibility of interatomic potentials used in the MD simulation. The neural-network (NN) interatomic potential is one of promising interatomic potentials based on machine-learning method. However, there are some parameters that should be determined heuristically before making the NN potential, such as the shape and number of basis functions. We have developed a new approach to select only relevant basis functions from a lot of candidates systematically and less heuristically without loosing the accuracy of the potential. The present NN potential for Si system shows very good agreements with the results obtained using ab-initio calculations.

scholarly journals Implementation of Deep Learning-based Automatic Modulation Classifier on FPGA SDR Platform

Electronics ◽  
2018 ◽  
Vol 7 (7) ◽  
pp. 122 ◽  
Author(s):  
Zhi-Ling Tang ◽  
Si-Min Li ◽  
Li-Juan Yu
Keyword(s):  
Neural Network ◽  
Radio Spectrum ◽  
Modulation Classification ◽  
Modulation Recognition ◽  
The Neural Network ◽  
Automatic Modulation Recognition ◽  
Field Programmable ◽  
Programmable Gate Arrays ◽  
On Chip ◽  
Automatic Modulation Classifier

Intelligent radios collect information by sensing signals within the radio spectrum, and the automatic modulation recognition (AMR) of signals is one of their most challenging tasks. Although the result of a modulation classification based on a deep neural network is better, the training of the neural network requires complicated calculations and expensive hardware. Therefore, in this paper, we propose a master–slave AMR architecture using the reconfigurability of field-programmable gate arrays (FPGAs). First, we discuss the method of building AMR, by using a stack convolution autoencoder (CAE), and analyze the principles of training and classification. Then, on the basis of the radiofrequency network-on-chip architecture, the constraint conditions of AMR in FPGA are proposed from the aspects of computing optimization and memory access optimization. The experimental results not only demonstrated that AMR-based CAEs worked correctly, but also showed that AMR based on neural networks could be implemented on FPGAs, with the potential for dynamic spectrum allocation and cognitive radio systems.

Neural Network for Automatic Analysis of Motility Data

1994 ◽  
Vol 33 (01) ◽  
pp. 157-160 ◽  
Author(s):  
S. Kruse-Andersen ◽  
J. Kolberg ◽  
E. Jakobsen
Keyword(s):  
Neural Network ◽  
Muscular Activity ◽  
Recording System ◽  
Automatic Analysis ◽  
Biologically Relevant ◽  
System A ◽  
The Neural Network ◽  
Valuable Method ◽  
Motor Abnormalities ◽  
Trained Network

Abstract:Continuous recording of intraluminal pressures for extended periods of time is currently regarded as a valuable method for detection of esophageal motor abnormalities. A subsequent automatic analysis of the resulting motility data relies on strict mathematical criteria for recognition of pressure events. Due to great variation in events, this method often fails to detect biologically relevant pressure variations. We have tried to develop a new concept for recognition of pressure events based on a neural network. Pressures were recorded for over 23 hours in 29 normal volunteers by means of a portable data recording system. A number of pressure events and non-events were selected from 9 recordings and used for training the network. The performance of the trained network was then verified on recordings from the remaining 20 volunteers. The accuracy and sensitivity of the two systems were comparable. However, the neural network recognized pressure peaks clearly generated by muscular activity that had escaped detection by the conventional program. In conclusion, we believe that neu-rocomputing has potential advantages for automatic analysis of gastrointestinal motility data.

Sign in / Sign up

Export Citation Format

Share Document