scholarly journals On the Reliability Assessment of Artificial Neural Networks Running on AI-Oriented MPSoCs

2021 ◽  
Vol 11 (14) ◽  
pp. 6455
Author(s):  
Annachiara Ruospo ◽  
Ernesto Sanchez
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Fault Tolerant ◽  
Optimal Solution ◽  
Computing System ◽  
Traditional Scheduling ◽  
Multiprocessor System ◽  
Critical Systems ◽  
Processing Elements ◽  
Artificial Neural

Nowadays, the usage of electronic devices running artificial neural networks (ANNs)-based applications is spreading in our everyday life. Due to their outstanding computational capabilities, ANNs have become appealing solutions for safety-critical systems as well. Frequently, they are considered intrinsically robust and fault tolerant for being brain-inspired and redundant computing models. However, when ANNs are deployed on resource-constrained hardware devices, single physical faults may compromise the activity of multiple neurons. Therefore, it is crucial to assess the reliability of the entire neural computing system, including both the software and the hardware components. This article systematically addresses reliability concerns for ANNs running on multiprocessor system-on-a-chips (MPSoCs). It presents a methodology to assign resilience scores to individual neurons and, based on that, schedule the workload of an ANN on the target MPSoC so that critical neurons are neatly distributed among the available processing elements. This reliability-oriented methodology exploits an integer linear programming solver to find the optimal solution. Experimental results are given for three different convolutional neural networks trained on MNIST, SVHN, and CIFAR-10. We carried out a comprehensive assessment on an open-source artificial intelligence-based RISC-V MPSoC. The results show the reliability improvements of the proposed methodology against the traditional scheduling.

The Faulty Behavior of Feedforward Neural Networks with Hard-Limiting Activation Function

1997 ◽  
Vol 9 (5) ◽  
pp. 1109-1126
Author(s):  
Zhiyu Tian ◽  
Ting-Ting Y. Lin ◽  
Shiyuan Yang ◽  
Shibai Tong
Keyword(s):  
Neural Networks ◽  
Monte Carlo ◽  
Artificial Neural Networks ◽  
Monte Carlo Simulations ◽  
Hardware Implementation ◽  
Fault Tolerant ◽  
Feedforward Neural Networks ◽  
Activation Function ◽  
Artificial Neural

With the progress in hardware implementation of artificial neural networks, the ability to analyze their faulty behavior has become increasingly important to their diagnosis, repair, reconfiguration, and reliable application. The behavior of feedforward neural networks with hard limiting activation function under stuck-at faults is studied in this article. It is shown that the stuck-at-M faults have a larger effect on the network's performance than the mixed stuck-at faults, which in turn have a larger effect than that of stuck-at-0 faults. Furthermore, the fault-tolerant ability of the network decreases with the increase of its size for the same percentage of faulty interconnections. The results of our analysis are validated by Monte-Carlo simulations.

scholarly journals Application of artificial neural networks in the process of drying CPC

2020 ◽  
Vol 216 ◽  
pp. 01128
Author(s):  
Yunusov Bakhtiar
Keyword(s):  
Neural Network ◽  
Experimental Data ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Artificial Neural Networks ◽  
Ethical Issues ◽  
Computing System ◽  
Powdered Cellulose ◽  
Neural Simulation ◽  
Artificial Neural

Training of artificial neural networks using the experimental data obtained in the process of drying of cotton powdered cellulose (CPC), have shown that, experiments can be conducted on the computers, which solves many of the practical and ethical issues. Therefore, arose and remain at the present time, the objective of neural simulation of processes and create a computing system (artificial neural network)

Multilayer Backpropagation Neural Networks for Implementation of Logic Gates

2021 ◽  
Vol 12 (1) ◽  
pp. 1-12
Author(s):  
Santosh Giri ◽  
Basanta Joshi
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Artificial Neural Networks ◽  
Specific Problem ◽  
Logic Gates ◽  
Backpropagation Algorithm ◽  
Processing Elements ◽  
Artificial Neural ◽  
Hidden Neurons

ANN is a computational model that is composed of several processing elements (neurons) that tries to solve a specific problem. Like the human brain, it provides the ability to learn from experiences without being explicitly programmed. This article is based on the implementation of artificial neural networks for logic gates. At first, the 3 layers Artificial Neural Network is designed with 2 input neurons, 2 hidden neurons & 1 output neuron. after that model is trained by using a backpropagation algorithm until the model satisfies the predefined error criteria (e) which set 0.01 in this experiment. The learning rate (α) used for this experiment was 0.01. The NN model produces correct output at iteration (p)= 20000 for AND, NAND & NOR gate. For OR & XOR the correct output is predicted at iteration (p)=15000 & 80000 respectively.

Development of ANN with Adaptive Connections by CE

2006 ◽  
pp. 71-93
Author(s):  
Julián Dorado ◽  
Nieves Pedreira ◽  
Mónica Miguelez
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Evolutionary Computation ◽  
Real World ◽  
Development Process ◽  
Multilevel System ◽  
Processing Elements ◽  
Artificial Neural ◽  
Real World Problems

This chapter presents the use of Artificial Neural Networks (ANN) and Evolutionary Computation (EC) techniques to solve real-world problems including those with a temporal component. The development of the ANN maintains some problems from the beginning of the ANN field that can be palliated applying EC to the development of ANN. In this chapter, we propose a multilevel system, based on each level in EC, to adjust the architecture and to train ANNs. Finally, the proposed system offers the possibility of adding new characteristics to the processing elements (PE) of the ANN without modifying the development process. This characteristic makes possible a faster convergence between natural and artificial neural networks.

NEURAL NETWORK MODEL DEVELOPMENT FOR DISEASE RATE MONITORING AND PREDICTION OF ANTIEPIDEMIC MEASURE EFFECTIVENESS

2020 ◽  
Vol 2020 (10) ◽  
pp. 42-50
Author(s):  
Nataliya Sukhanova
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Network Model ◽  
Neural Network Model ◽  
Model Development ◽  
Computing System ◽  
Disease Rate ◽  
Artificial Neural ◽  
Visualization Of Data

There is developed a neural network model for disease rate prediction and assessment of antiepidemic measure effectiveness. As basis of the development there were adopted the existing automated information systems which are used for monitoring and visualization of data on Moscow population disease rate. Under conditions of the emergence and propagation of new dangerous infectious and virus diseases the information processing must be carried out in real time, a prediction for future is required. It is necessary to create, update and adjust rapidly a set of anti-epidemic measures offered. The investigation purpose consists in the prediction of infection spreading and the assessment of anti-epidemic measures based on data on the population disease rate. There is offered a neural network model realized on the basis of the modular computing system and artificial neural networks. A modular computing system includes modules of different types connected between each other with a switch network. In the modular computing system there are included modules of artificial neural networks with the special switch structure. Switchboards allow connecting and disconnecting single modules and elements of neural networks. A neural network model changes dynamically its structure and adapted to a current epidemic situation.

Sign in / Sign up

Export Citation Format

Share Document